Key Points
Question
Are cryopreserved platelets less effective and safe than conventional liquid-stored platelets in the treatment of cardiac surgical bleeding?
Findings
In bleeding adult cardiac surgery patients, cryopreserved platelets did not significantly increase 24-hour postoperative blood loss, although the plausible upper limit of blood loss exceeded the prespecified 20% noninferiority margin, and cryopreserved platelets were associated with significantly greater intraoperative and total blood loss as well as significantly more transfusions of other blood components. Despite a similar incidence of prespecified adverse events, patients receiving cryopreserved platelets were extubated and discharged from the intensive care unit and hospital significantly later.
Meaning
Cryopreserved platelets did not meet the criterion for noninferiority to liquid-stored platelets for bleeding control in cardiac surgery.
Abstract
Importance
Liquid-stored platelets have a shelf-life of 5 to 7 days, limiting availability and resulting in wastage.
Objective
To assess the effectiveness and safety of dimethyl sulfoxide–cryopreserved platelets, which have a shelf-life of 2 years, as a treatment for cardiac surgery bleeding.
Design, Setting, and Participants
The Cryopreserved vs Liquid Platelets II (CLIP-II) trial was a multicenter, randomized, double-blind, parallel-group noninferiority trial, which enrolled patients between August 2021 and April 2024 at 11 Australian tertiary hospitals, with follow-up completed in July 2024. Patients at high risk of platelet transfusion were eligible. Patients were excluded if they had a history of deep vein thrombosis or pulmonary embolism, were coagulopathic, or were females aged 18 to 55 years who were rhesus D (RhD) negative or of unknown RhD status. Of 879 patients meeting inclusion criteria, 182 were excluded and 285 did not consent, leaving 412. Of these, 388 were randomized and 202 received study platelets.
Interventions
Patients received up to 3 units of either group O cryopreserved platelets or conventional liquid-stored platelets, commencing intraoperatively or in the first 24 postoperative hours.
Main Outcomes and Measures
The primary outcome was postsurgical chest drain bleeding within the first 24 hours following intensive care unit admission. Noninferiority was defined prospectively as less than 20% greater bleeding in this period. Five secondary and 42 tertiary outcomes were defined a priori.
Results
Of the 202 transfused patients (mean [SD] age, 64.4 [13] years; 75.7% male), 61 (30.2%) underwent nonelective surgery. The primary outcome did not differ between groups (605 mL in cryopreserved platelet group vs 535 mL in liquid-stored platelet group; ratio of geometric means [cryopreserved to liquid ratio], 1.13 [95% CI, 0.96-1.34]; P = .07). As the confidence interval includes bleeding exceeding the noninferiority margin, noninferiority was not established. Cryopreserved platelet transfusion was associated with higher intraoperative and total perioperative blood loss (ratio of geometric means [cryopreserved to liquid ratio], 1.42 [95% CI, 1.12-1.80]; 1.31 [95% CI, 1.07-1.60], respectively), and increased red cell, plasma, and cryoprecipitate transfusion. While there were no differences in the incidence of prespecified adverse events, patients receiving cryopreserved platelets experienced longer times to extubation and intensive care unit/hospital discharge (median [IQR] duration of ventilation, 25.5 hours [16.1-77.3] vs 23.6 hours [13.1-52.8]; median [IQR] intensive care unit length of stay, 3.8 days [2.0-6.0] vs 3.0 days [1.9-4.9]; median hospital length of stay, 10.9 days [7.87-17.0] vs 9.1 [6.9-14.9]).
Conclusions and Relevance
Cryopreserved platelets did not meet the predefined threshold for noninferiority in hemostatic effectiveness at 24 hours after ICU admission. Additional predefined end points consistently indicated diminished hemostatic effectiveness, although prespecified adverse events were comparable.
Trial Registration
ClinicalTrials.gov Identifier: NCT03991481
This clinical trial compares the effectiveness and safety of dimethyl sulfoxide–cryopreserved platelets vs conventional liquid-stored platelets as a treatment for cardiac surgery bleeding.
Introduction
Platelets are essential in the balanced transfusion response to major hemorrhage, catalyzing thrombin generation and the formation of mechanical clot scaffolds. International guidelines1,2 recommend early platelet transfusion to reduce mortality.
Conventionally, platelets are stored at 22 °C to optimize posttransfusion circulation. Bacterial growth limits storage duration to only 5 to 7 days leading to wastage, with 25% to 33% of donated platelet units discarded.3,4 Typically, only larger hospitals with high and predictable demand carry large stocks of platelets. Smaller, remote, and military hospitals usually have none.5
Fifty years ago, the US Navy reported using dimethyl sulfoxide (DMSO) to cryopreserve platelets at −80 °C,6 extending their shelf-life to 2 years. This technique was later modified, removing DMSO prior to freezing7 and resuspending platelets in plasma rather than saline.8 Resuspension takes approximately 30 minutes and requires minimal equipment and training. Preclinical studies have found altered receptor expression, enhanced agonist reactivity, and more platelet microparticles with cryopreservation.9 Limited case series,8 safety studies,10 nonrandomized comparative studies,11 and 3 small randomized trials in bleeding patients12,13,14 have suggested greater hemostatic activity without adverse effects. However, evidence has been insufficient for regulatory approval.
Accordingly, the effectiveness and safety of cryopreserved platelets, compared with liquid-stored platelets, as treatment for bleeding were assessed in this study. A noninferiority design was chosen due to the potential advantage in availability associated with cryopreserved platelets. While liquid-stored platelets are standard care for this indication, no clinical trial has established the efficacy of this reference treatment. Cardiac surgery patients were studied, as cardiac surgery is the most common cause of bleeding in hospitalized patients, accounting for 44% of surgical platelet use.4 In Australia, 24% of cardiac surgical patients receive platelet transfusions.15 Patients at higher risk can be identified preoperatively.16 Chest drain blood loss can be quantified accurately, and cardiac surgical patients can provide prospective consent.
Methods
The Cryopreserved vs Liquid Platelets II (CLIP-II) study was an investigator-initiated, randomized, double-blind, multicenter, parallel-group, noninferiority trial conducted in 11 Australian hospitals between August 25, 2021, and April 16, 2024. Patient follow-up was completed in July 2024. The trial was overseen by a management committee on behalf of 3 trial groups. The trial protocol and investigator brochure (Supplements 1 and 2) were approved by the Austin Health and Australian Red Cross Lifeblood Human Research Ethics Committees. This study is reported according to Consolidated Standards of Reporting Trials (CONSORT) guidelines.
Patients
Patients 18 years and older at high risk of requiring a platelet transfusion were identified using the Australian Cardiac Surgery Platelet Transfusion (ACSePT) score (Figure 1).16 Approximately 30% of cardiac surgery patients have an ACSePT score of 1 or greater, corresponding to a more than 30% probability of platelet transfusion. Patients were also eligible if their clinicians believed they were likely to require platelets. Patients were excluded if they had been previously enrolled in a trial of a bleeding medication or if they had a history of deep vein thrombosis or pulmonary embolism. As all study cryopreserved platelets were collected from rhesus D (RhD)–positive donors, females between 18 and 55 years of age who were RhD negative or had an unknown RhD status were excluded. Patients with abnormal coagulation, due either to bleeding conditions or medications, were also excluded because coagulopathies could plausibly have had a greater effect on the primary outcome than any difference in the effectiveness of the trial interventions. Written consent was sought at least 1 day prior to planned surgery by investigators at each hospital.
Figure 1. Recruitment, Randomization, and Patient Flow in the CLIP-II Triala.
CLIP-II indicates Cryopreserved vs Liquid Platelets II; DVT, deep vein thrombosis; and PE, pulmonary embolism.
aData on the number of patients screened but who did not meet the inclusion criteria were not collected in this study.
bThe Australian Cardiac Surgery Platelet Transfusion (ACSePT) score ranges from −0.5 to 5. An ACSePT score of ≥1 is associated with a 30% incidence of platelet transfusion in the perioperative period.
cMultiple exclusion reasons were possible.
Interventions
Cryopreserved group O apheresis platelets were reconstituted in ABO-group matched or AB plasma using the Netherlands Military Blood Bank method,8 with slight modifications.9,17 Liquid platelets in a mixture of plasma and platelet additive solution were either derived from whole blood (pooled from the buffy coats from 4 donors) or donated by apheresis, with ABO group matching according to hospital practice.
A computer-generated permuted block randomization schedule, generated by the study coordinating center, with random block sizes of 2 or 4 and allocation concealment, was used to randomly assign patients 1:1 to either cryopreserved or liquid-stored platelets. Patients received study platelets only when their clinicians judged transfusion necessary, with decisions made at clinicians’ discretion in line with usual Australian practice. Each patient was given up to 3 study platelet units, after which open-label liquid-stored platelets were administered if indicated. The first study platelet unit was commenced either intraoperatively or within the first 24 hours following intensive care unit (ICU) admission. The remaining 2 study platelet units could be commenced at any time until ICU discharge.
Study platelets were issued from the blood bank with an opaque cover. Two unblinded nurses at each hospital, not involved in the patients’ care, conducted routine pretransfusion checks of patient identity, ABO and RhD group, product integrity, and expiry date.17 Participants, clinicians administering the interventions, and outcome assessors remained blinded to group allocation throughout the trial.
Outcomes
The primary outcome was postsurgical chest drain bleeding volume within the first 24 hours following ICU admission. Secondary and tertiary effectiveness and safety outcomes were prospectively defined.17
The trial protocol (Supplement 1) and statistical analysis plan17 were published prior to completion of the interim analysis, conducted at 50% recruitment. Minor modifications are presented in Supplement 3. Transfusion of 202 patients conferred greater than 80% power (1 -sided P value of .025) to detect a noninferior limit in the primary outcome of 20%. These calculations included 15% inflation for possible nonnormality18 and were derived from our pilot study13 in which the observed mean (SD) blood loss within 24 hours after ICU admission was 943 (426) mL. A 20% noninferior limit was chosen as the mean volume of a red cell unit for transfusion in Australia is 258 mL,19 slightly more than 20% of the anticipated primary outcome blood loss.
Randomized patients who never required a platelet transfusion and those who received no study platelet units due to inadequate personnel in hospital blood banks after hours were excluded from comparative analysis. The primary analysis included all randomized patients who received at least 1 study platelet unit. Sensitivity to adherence was analyzed by excluding patients who received open-label platelets prior to receiving all 3 study platelet units without meeting the withholding criteria. The sensitivity analysis was conducted for the primary and all secondary bleeding outcomes.
Several prespecified subgroup analyses were performed. For reasons detailed in the eAppendix in Supplement 3, a subgroup analysis compared only group O recipients in the cryopreserved group with only ABO-type–matched recipients in the liquid platelets–stored group. Subgroups defined by complexity of surgery, preoperative aspirin usage, and location of first platelet transfusion were also analyzed.
Statistical Analysis
Given the log-normal distribution of patient blood loss, both the primary outcome and blood loss secondary outcome data were log-transformed. Unadjusted comparisons were performed using t tests and adjusted analysis used log-linear mixed modeling, with results presented as geometric means and ratio of geometric means (cryopreserved to liquid ratio). For increased transparency, medians (IQRs) have also been reported. Binary outcomes were analyzed using Poisson regression with an identity link function to derive risk differences, using robust standard errors. Nonparametric outcomes were analyzed using median regression with results reported as a difference of medians. Time-to-event outcomes were analyzed using Fine and Gray models to account for the competing risk of death with results reported as subdistribution hazard ratios and presented as cumulative incidence plots. Multivariable analysis was performed for all primary and secondary outcomes adjusting for site and patient risk (EuroSCORE II20). Multiple imputation was used to assess sensitivity to missingness, which was rare (n = 1).
Statistical significance for the primary outcome was determined by a 1-sided P value of .025. All results are reported with 95% confidence intervals. In the absence of adjustment for multiplicity, confidence intervals for secondary and tertiary outcomes should not be used in place of hypothesis testing. Supplement 3 provides additional details on the analyses, which were performed using SAS version 9.4 (SAS Institute).
Results
Patient Characteristics
Of the 879 patients who met inclusion criteria (Figure 1), 182 were excluded and consent was not obtained from 285. Of the 412 consenting patients, 388 were randomized and of those, 202 were deemed to require a platelet transfusion (104 received cryopreserved platelets and 98 received liquid-stored platelets). Trial groups were well matched at baseline (Table 1; eTable 1 in Supplement 3), with several indices reflecting high bleeding risk: more than 30% nonelective surgery, more than 60% abnormal kidney function, long21 mean (SD) cardiopulmonary bypass time (188 [85.8] minutes), and median (IQR) EuroSCORE II of 3.4 (2.2-7.0). Most (>70%) were recruited from 3 hospitals (eTable 2 in Supplement 3).
Table 1. Characteristics of Patients at Baseline in the CLIP-II Trial.
| Characteristic | Cryopreserved platelets (n = 104) | Liquid-stored platelets (n = 98) |
|---|---|---|
| Demographics | ||
| Age, mean (SD), y | 65.5 (12.0) | 63.2 (14.0) |
| Sex, No. (%) | ||
| Male | 81 (77.9) | 72 (73.5) |
| Female | 23 (22.1) | 26 (26.5) |
| BMI, mean (SD) | 28.5 (5.9) | 27.4 (6.3) |
| EuroSCORE II risk score, median (IQR)a | 3.5 (2.1-7.7) | 3.4 (2.2-6.2) |
| Nature of surgery, No. (%) | ||
| Elective surgery | 73 (70.2) | 68 (69.4) |
| Deep hypothermic arrestb | 17 (16.3) | 17 (17.3) |
| Minimally invasive operation | 1 (1.0) | 4 (4.1) |
| Total cardiopulmonary bypass time, mean (SD), min | 195 (94) [n = 103] | 180 (75) |
| Laboratory test results, No. (%) | ||
| Creatinine clearance >85 mL/min | 40 (38.5) | 41 (41.8) |
| Preoperative creatinine >1.13 mg/dL | 38 (36.5) | 27 (27.6) |
| Preoperative hemoglobin <100 g/dL | 12 (11.5) | 12 (12.2) |
| Preoperative platelet count <100 × 103/µL | 1 (1.0) | 2 (2) |
| Preoperative aPTT >40 s | 8 (7.9) [n = 101] | 12 (12.6) [n = 95] |
| Preoperative fibrinogen <200 mg/dL | 0 [n = 91] | 1 (1.1) [n = 87] |
| Preoperative INR >1.5 | 3 (3) [n = 101] | 2 (2.1) [n = 95] |
| Preoperative treatments, No. (%) | ||
| Tranexamic acid in the 24 h prior to surgery | 8 (7.7) | 7 (7.1) |
| Any oral, intravenous, or subcutaneous anticoagulant therapy in 7 d prior to the surgery | 62 (59.6) | 58 (59.2) |
| Aspirin in 7 d prior to the surgery | 34 (32.7) | 36 (36.7) |
| Prophylactic enoxaparin in 7 d prior to the surgery | 17 (16.3) | 27 (27.6) |
| Prophylactic heparin in 7 d prior to the surgery | 17 (16.3) | 10 (10.2) |
Abbreviations: aPTT, activated partial thromboplastin time; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CLIP-II, Cryopreserved vs Liquid Platelets II; INR, international normalized ratio.
SI conversion factors: To convert creatinine to µmol/L, multiply by 88.4; and creatinine clearance to mL/s/m2, multiply by 0.0167.
eTable 1 in Supplement 1 presents all the individual components of this score. EuroSCORE II calculates the predicted hospital mortality for cardiac surgery patients based on 18 preoperative variables. Possible values range from 0.5% to 99.78%, with overall mortality in the original 22 381-patient study of 3.9%.
Deep hypothermic arrest involves systemic cooling to 18-20 °C during cardiopulmonary bypass to provide cerebral protection while circulatory arrest is established to facilitate surgical access.
Study Intervention
Most patients (86.6%) were first transfused study platelets in the operating room; this did not differ between groups (Table 2). Despite the possibility that thawing plasma might delay receipt of cryopreserved platelets, the time between ordering and receiving cryopreserved vs liquid-stored platelets was nearly identical. More patients in the cryopreserved group (25% vs 3.1%; difference, 21.9% [95% CI, 12.9% to 30.9%]) received open-label platelets after the first study platelet unit but before all 3 study platelets had been transfused. The total number of study plus open-label platelets was higher in the cryopreserved platelets group (median [IQR], 2 [1 to 4] vs 1 [1 to 2] units; difference, 1 [95% CI, 1 to 1] unit). Despite this, the 24-hour platelet nadir was lower in the cryopreserved platelets group (median [IQR], 103 [81 to 129] vs 140 [110 to 177] ×103/µL; difference, −36 [95% CI, −50 to −22]) ×103/µL). In the cryopreserved platelets group but not the liquid-stored platelets group, platelet nadirs were further reduced among those who received open-label platelets (eTable 9 in Supplement 3).
Table 2. Study Intervention and Effectiveness Outcomes.
| Cryopreserved platelets (n = 104) | Liquid-stored platelets (n = 98) | Ratio of geometric means and difference of proportions (95% CI) or medians (IQR) | ||
|---|---|---|---|---|
| Absolute difference | Adjusted treatment effecta | |||
| Primary outcome | ||||
| Postsurgical bleeding in the first 24 h in ICU, mL | ||||
| Geometric mean (95% CI) | 605 (532 to 688) [n = 103] | 535 (480 to 596) | 1.13 (0.96 to 1.34) | 1.11 (0.94 to 1.31)b,c |
| P value | .07b | |||
| Median (IQR) | 610 (410 to 950) [n = 103] | 523 (370 to 780) | 87 (−50 to 280) | 99 (−16 to 215) |
| Secondary outcomes | ||||
| Total volume of postsurgical chest drain bleeding, mL | ||||
| Geometric mean (95% CI) | 1273 (1096 to 1479) [n = 103] | 972 (848 to 1114) | 1.31 (1.07 to 1.60) | 1.29 (1.06 to 1.56) |
| Median (IQR) | 1210 (750 to 2110) [n = 103] | 1058 (562 to 1570) | 152 (−70 to 590) | 311 (29 to 592) |
| Composite bleeding outcome using the BARC4 criteria, No. (%)d | 33 (31.7) | 19 (19.4%) | 12.3 (0.5 to 24.2) | 9.8 (−0.6 to 20.3) |
| RBC transfused in the first 24, No. (%) | 55 (52.9) | 38 (38.8) | 14.1 (0.5 to 27.7) | 13.2 (3.8 to 22.5) |
| Units of RBC transfused in the first 24 h post ICU admission, median (IQR) | ||||
| All patients | 1 (0 to 3) | 0 (0 to 1) | 1 (0 to 2) | 0 (0 to 1) |
| Only those transfused | 3 (1 to 4) | 2 (1 to 3) | 1 (0 to 2) | 1 (0 to 2) |
| RBC transfusion received at any point, No. (%) | 77 (74.0) | 58 (59.2) | 14.9 (2.0 to 27.7) | 13.8 (1.7 to 26.0) |
| Total units of red blood cells transfused by the time of ICU discharge or day 7, median (IQR) | ||||
| All patients | 3 (0 to 7) | 2 (0 to 4) | 1 (−1 to 3) | 1 (0 to 2) |
| Only those transfused | 4 (2 to 9) | 4 (2 to 6) | 0 (−2 to 2) | 1 (−1 to 2) |
| Select tertiary outcomes | ||||
| FFP transfusion received at any point, No. (%) | 78 (75.0) | 49 (50.0) | 25.0 (12.1 to 37.9) | |
| Cryoprecipitate transfusion received at any point, No. (%) | 74 (71.2) | 53 (54.1) | 17.1 (3.9 to 30.2) | |
| Lowest platelet count–first 24 h in ICU, median (IQR), ×103/µL | 103 (81 to 129) [n = 103] | 140 (110 to 177) | −36 (−50 to −22) | |
| Lowest hemoglobin–first 24 h in ICU, median (IQR), g/dL | 83 (74 to 94) | 88.5 (78 to 100) | −5 (−11 to 1) | |
| Study intervention | ||||
| First platelets in operating room, No. (%) | 88 (84.6) | 87 (88.8) | −4.2 (−13.5 to 5.2) | |
| Delay between platelet order and commencement of first study platelet infusion, median (IQR), min | 60 (36 to 109) [n = 102] | 57 (33 to 100) | 2 (−15 to 19) | |
| No. of units of study platelets received, median (IQR), No. | 2 (1 to 2) | 1 (1 to 2) | 1 (1 to 1) | |
| Received open-label units before any study platelets, No. (%) | 1 (1.0) | 0 | 1.0 (−0.9 to 2.8) | |
| Received open-label units after first unit of study platelets but before all 3 study units were completed, No. (%) | 26 (25.0) | 3 (3.1) | 21.9 (12.9 to 30.9) | |
| Received open-label units after 3 units of study platelets, No. (%) | 17 (16.3) | 9 (9.2) | 7.2 (−2.0 to 16.3) | |
| Received open-label units at any time, No. (%) | 44 (42.3) | 12 (12.2) | 30.1 (18.6 to 41.6) | |
| Total No. of open-label units received, median (IQR) | 2 (1 to 4) | 1 (1 to 3) | 1 (0 to 2) | |
| Total No. of all platelets received [study + open label], median (IQR) | 2 (1 to 4) | 1 (1 to 2) | 1 (1 to 1) | |
Abbreviations: BARC4, Bleeding Academic Research Consortium type 4; FFP, fresh frozen plasma; ICU, intensive care unit; RBC, red blood cells.
Adjusted for site and EuroSCORE II.
1-sided P value.
Following multiple imputation of a single missing patient, the adjusted treatment effect remained identical to 2 decimal places (ratio of geometric means [cryopreserved to liquid ratio], 1.11 [95% CI, 0.94-1.31]).
BARC4 bleeding outcome: any of intracranial bleeding within 48 hours; reoperation after closure of sternotomy; transfusion of ≥5 units of whole blood or red blood cells within the 48-hour intraoperative or postoperative period; or chest tube output ≥2 L within a 24-hour period.22
The mean (SD) age of liquid-stored platelet units transfused was 5.5 (1.0) days. Few liquid-stored units (15.6%) were obtained by apheresis; most were derived from whole blood. All cryopreserved platelet units were frozen in DMSO within 50 hours of apheresis donation, reconstituted within 2 years of freezing, and transfused within 4 hours of resuspension.
Primary Outcome
The study primary outcome, blood loss in the chest drains over the first 24 hours in the ICU, was not different between the study groups (median [IQR], 610 [410 to 950] vs 523 [370 to 780]; difference, 87 [−50 to 280]; ratio of geometric means, 1.13 [95% CI, 0.96 to 1.34]; P = .07). As the confidence interval extended beyond the prespecified noninferiority threshold, noninferiority of cryopreserved platelets could not be concluded (Figure 2).
Figure 2. Ratio of Geometric Means for the Study Primary Outcome.
The black dashed line marks the noninferiority margin at 1.2; the shaded region indicates values where cryopreserved platelets would be considered noninferior. Adjusted estimates account for site and EuroSCORE II.
Secondary and Tertiary Outcomes
Other prespecified study effectiveness outcomes were worse in the cryopreserved platelets group (Table 2; eTable 3 in Supplement 3). Intraoperative blood loss (ratio of geometric means, 1.42 [95% CI, 1.12-1.80]), chest drain volume at ICU admission (ratio of geometric means, 2.48 [95% CI, 1.40-4.38]), and total postoperative blood loss (ratio of geometric means, 1.31 [95% CI, 1.07-1.60]) were all higher in the cryopreserved group (eFigure 1 in Supplement 3). The Bleeding Academic Research Consortium type 4 (BARC4) composite bleeding outcome22 (any of intracranial bleeding within 48 hours; reoperation after closure of sternotomy; transfusion of ≥5 units of whole blood or red blood cells within the 48-hour intraoperative or postoperative period; chest tube output ≥2 L within a 24-hour period) occurred more frequently in the cryopreserved platelets group (cryopreserved platelets group, 31.7% vs liquid-stored platelets group, 19.4%; difference, 12.3% [95% CI, 0.5%-24.2%]).
More patients who received cryopreserved platelets were transfused red cells (cryopreserved platelets group, 74.0% vs liquid-stored platelets group, 59.2%; difference, 14.9% [95% CI, 2.0% to 27.7%]). The 24-hour hemoglobin concentration nadir was not lower in the cryopreserved group (median [IQR], cryopreserved platelets group, 83 [74 to 94] vs liquid-stored platelets group, 88.5 [78 to 100] mg/dL; difference, −5 [−11 to 1]). Fresh frozen plasma (FFP) and cryoprecipitate were transfused to a greater proportion of patients in the cryopreserved group (Table 2; eFigure 2 in Supplement 3). Other hemostatic treatments were uncommon and did not differ between groups (eTable 3 in Supplement 3). Inotrope and vasopressor requirements were not significantly different. Patients in the liquid-stored group received more intravenous crystalloid in the first 6 hours of their ICU stay (eTable 3 in Supplement 3).
Clinicians remove chest drains and commence prophylactic heparin and aspirin when they perceive risk of substantial bleeding has passed. While the time to start aspirin (eFigure 3 in Supplement 3) and heparin (eFigure 4 in Supplement 3) did not differ between study groups, a greater proportion of patients in the liquid-stored platelets group had their surgical drains removed at or before 24 hours (eTable 3 in Supplement 3).
Limited viscoelastic testing data were obtained after the last study platelet unit transfusion (eTable 4 in Supplement 3). Few differences in these parameters were observed, although the low numbers in each group preclude any definitive conclusion.
Safety
There were no differences in the prespecified (Table 3) or additional (eTable 5 in Supplement 3) adverse effects. Local and systemic infections were uncommon (1.5% and 3%, respectively), and few patients (3.5%) ever had a core temperature greater than 39 °C. While the mean lowest Pao2 to Fio2 (partial pressure of arterial oxygen to fraction of inspired oxygen) ratio in both study groups was low (215 in the cryopreserved platelets group and 218 in the liquid-stored platelets group) and the duration of mechanical ventilation was long (median, 24.9 hours), few patients developed acute respiratory distress syndrome23 (<2% in each group). A substantial proportion (13.4% overall) required kidney replacement therapy, but this did not differ between groups. Very few patients had arterial or venous thrombosis. Patients who received liquid-stored platelets were extubated sooner and discharged from the ICU and hospital earlier (eFigures 5 to 7 in Supplement 3). While mortality did not differ between groups (Table 3; eFigure 8 in Supplement 3), numerically more patients in the cryopreserved platelets group died in the ICU, hospital, and by day 90.
Table 3. Safety Outcomes.
| Safety events | Cryopreserved platelets (n = 104)a | Liquid-stored platelets (n = 98)a |
|---|---|---|
| Died, No. (%) | ||
| In ICU | 9 (8.7) | 3 (3.1) |
| In hospital | 11 (10.6) | 4 (4.1) |
| At day 90 | 13 (12.5) | 5 (5.1) |
| Reoperation after closure of sternotomy for control of bleeding, No. (%) | 9 (8.7) | 7 (7.1) |
| Acute respiratory distress syndrome, No. (%) | 2 (1.9) | 1 (1.0) |
| Systemic infection, No. (%) | 5 (4.8) | 1 (1.0) |
| Local sternal wound infection, No. (%) | 3 (2.9) | 0 |
| Temperature >39 °C at any time in ICU stay, No. (%) | 5 (4.9) | 2 (2.0) |
| Pulmonary embolus, No. (%) | 1 (1.0) | 2 (2.0) |
| Peripheral arterial occlusion, No. (%) | 1 (1.0) | 3 (3.1) |
| Total duration of mechanical ventilation, median (IQR), h | 25.5 (16.1-77.3) | 23.6 (13.1-52.8) |
| ICU length of stay (first admission), median (IQR), d | 3.8 (2.0-6.0) | 3.0 (1.9-4.9) |
| Hospital length of stay from first platelet transfusion, median (IQR), d | 10.9 (7.87-17.0) | 9.1 (6.9-14.9) |
Abbreviation: ICU, intensive care unit.
No occurrences of acute coronary syndrome, deep vein thrombosis, or prespecified dimethyl sulfoxide complications (nausea, headache, tachycardia, bradycardia, hypertension) were observed in either group.
Adjusted and Subgroup Analyses
When adjusted for hospital site and patient risk (EuroSCORE II), there was no qualitative change in between-group differences in the primary outcome or any secondary outcomes (Table 2).
There was no significant heterogeneity in the primary outcome in any of the 4 predefined subgroups (transfused ABO-compatible platelets, complex surgery, preoperative aspirin usage, and location of first study platelet transfusion; eFigure 9 in Supplement 3).
Twenty-nine patients received open-label platelets prior to transfusion of all 3 units of study platelets when no withholding criteria were met (Figure 1). In a sensitivity analysis that included only patients who received the protocol as intended (eTables 6-8 in Supplement 3), there remained no difference in the primary outcome of 24-hour ICU blood loss (geometric mean, cryopreserved platelets group, 571 [95% CI, 493-662] mL vs liquid-stored platelets group, 535 [95% CI, 479-598] mL; ratio of geometric means, for cryopreserved to liquid ratio, 1.07 [95% CI, 0.89-1.28]). Comparisons of other study outcomes in the sensitivity analysis remained consistent with the primary analysis findings.
The prespecified comparison of total estimated health care costs and cost-effectiveness in the 2 study groups will be reported in a subsequent publication.
Discussion
Cryopreserved platelets did not demonstrate noninferiority in 24-hour postoperative ICU blood loss when compared with liquid-stored platelets, with confidence intervals that indicated a potential increase in blood loss of up to 34%. Several congruent prespecified secondary outcomes including blood loss during other time periods, transfusion of blood components, time of chest drain removal, and time to extubation, and time to ICU and hospital discharge all suggested that cryopreserved platelets were less hemostatically effective. Cryopreserved platelets were associated with a low incidence of adverse events that did not differ from that of liquid-stored platelets. While the trial was powered for these prespecified events, it cannot exclude more subtle or unexpected safety signals. Further, while not significantly different, mortality outcomes were numerically worse in the cryopreserved group.
Previous smaller and preclinical studies have proposed that cryopreserved platelets exhibit a preactivated phenotype, leading to more rapid circulatory clearance. While bleeding hematology patients had dose-dependent increases in platelet counts and all patients’ bleeding was stabilized, cryopreserved platelet transfusion caused a significantly lower increment in platelet count than did liquid-stored platelet transfusion.10 This lesser platelet increment was also observed in previous pilot trials,13,14 the earlier phase 2b trial,12 and a nonrandomized comparison in trauma patients,11 consistent with the result of the current study.
The earlier phase 2b clinical trial randomized 73 patients to cryopreserved or liquid-stored platelets to treat bleeding after cardiac surgery,12 reporting no adverse effects in the 24 patients who received cryopreserved platelets. Unlike the current study, blood loss and transfusion requirement associated with cryopreserved platelets was significantly less. The reason for these differences is speculative. The method of cryopreserved platelet preparation and reconstitution was different: in that trial, platelets were stored in a larger volume of DMSO/plasma, necessitating washing in a 0.9% saline/glucose/phosphorus solution immediately prior to reconstitution. Effects of these differences are not known. Methodological differences in analysis might be relevant. In the Khuri et al12 trial, “surgical mishaps…were prospectively identified, and affected patients were excluded from analysis.” Numbers excluded for this reason were not reported. In contrast, the current trial did not exclude patients based on judgments regarding the nature of bleeding, thereby avoiding postrandomization selection bias. In the current trial, 13% of first study platelet transfusions occurred in the ICU, so some postoperative bleeding would not have been amenable to the study intervention. This proportion was not reported in the Khuri et al trial.12
The dose equivalence of cryopreserved vs liquid platelets is uncertain. Quality control results from 81 randomly selected cryopreserved platelet units manufactured during the trial found a mean (SD) of 341 (39) ×109 platelets per unit (unpublished data, D.C.M., Australian Red Cross Lifeblood, October 2025), somewhat higher than the 287 (35) ×109 per apheresis unit and 264 (37) ×109 per pooled platelet unit19 used in the trial liquid-stored platelets group. However, the phenotype of cryopreserved platelets differs substantially,24 suggesting platelet count alone is not adequate to assure dose equivalence. For hypothesized safety reasons, patients were limited to 3 study platelet units, limiting the ability to comment on whether greater quantities of cryopreserved platelets might have improved hemostasis. Future studies may build on the current findings of a low and comparable prespecified adverse event rate to more confidently investigate this question.
This study’s strengths include its multicenter, randomized, blinded design that used standard care for comparison. Outcomes were measured objectively. The pragmatic design relied on physicians titrating platelet transfusion to effect, rather than an artificial protocol. The concern that cryopreserved platelets might be delayed due to plasma thawing was unfounded. Even had a time difference been observed, this would have been inherent in the thawing process and so a necessary element of the evaluation.
A pilot trial reported substantially higher 24-hour blood loss after ICU admission (mean [SD], 943 [426] mL) than was observed in CLIP-II (mean [SD], 676 [428] mL). This is likely to reflect differences in patient selection (TRUST25 vs ACSePT16 score) and possibly temporal improvements in perioperative management. While the variability underpinning the power calculation was preserved, the lower observed bleeding left the trial underpowered to exclude the prespecified noninferiority margin. Approximately 87% of patients received their first study platelet transfusion intraoperatively, with a 42% (95% CI, 12%-80%) higher intraoperative blood loss in the cryopreserved group. Quantifying platelet transfusion effects sooner after administration, when bleeding was heaviest, might have been more representative. However, intraoperative blood loss estimation is often inaccurate,26 and these data were recorded in only two-thirds of study patients. Further, intraoperative blood loss can result primarily from causes amenable only to surgical control rather than from coagulopathic causes. Consequently, the greater intraoperative hemorrhage in the cryopreserved platelets group might be misleading. There were negligible missing data (<0.5%) for total recorded postoperative blood loss, which is likely to be more accurate and should reflect coagulopathic rather than surgical bleeding. This bleeding outcome was 31% (95% CI, 7%-60%) higher with cryopreserved platelets.
Limitations
This study has other limitations. First, results in high-risk cardiac surgery patients might not translate to trauma, gastrointestinal bleeding, or obstetric hemorrhage. The lower platelet increments this study and others have observed suggest a different study design is required to evaluate prophylactic use in thrombocytopenic patients. Second, cryopreserved platelets were reconstituted in fresh frozen plasma while liquid-stored platelets were suspended in platelet additive solution with less plasma. This should favor a greater hemostatic effect in the cryopreserved platelets group; however, this study observed the opposite result. Third, a higher proportion of patients in the cryopreserved platelets group received open-label platelets before all 3 study platelets had been transfused, suggesting that clinicians possibly anticipated delayed receipt of cryopreserved platelets, or observed lesser hemostatic effectiveness or inadequate platelet increment. On average, cryopreserved platelets were not supplied with a longer delay, but the second and third of these possibilities potentially imply lack of blinding. Regardless, premature receipt of open-label liquid-stored platelets in the cryopreserved platelets group favors finding no outcome difference between trial groups; differences were observed despite this.
These results suggest that cryopreserved platelets should not replace liquid-stored platelets as a treatment for bleeding in situations where liquid-stored platelets are readily available. However, noting absence of hypothesized prespecified adverse events, cryopreserved platelets might still have utility in mitigating exhaustion of liquid platelet supply, as a means of providing human platelet antigen– and human leukocyte antigen–matched platelets,27 and as a means of providing platelets to small, rural, and military hospitals.28
Conclusions
Cryopreserved platelets did not meet criterion for noninferiority to liquid-stored platelets in treating bleeding during and after cardiac surgery.
Trial Protocol
Investigator Brochure
Trial Supplementary Information
eTable 1. Additional Baseline Data
eTable 2. Study Hospitals
eTable 3. Predefined Tertiary and Additional Effectiveness Outcomes
eTable 4. Viscoelastic Testing Results After the Last Study Platelet Transfusion
eTable 5. Additional Safety Outcome Data
eTable 6. Baseline Characteristics of Patients in the Sensitivity Analysis (Excluding Recipients of Open Label Prior to 3 Study Units)
eTable 7. Characteristics of Included vs Excluded Patients (Received Open Label Prior to 3 Study Units)
eTable 8. Primary and Secondary Outcomes for The Sensitivity Analysis (Excluding Patients That Received Open Label Prior to 3 Study Units (n=29))
eTable 9. Mixed-Effects Model Estimates of Lowest Platelet Count in the First 24 Hours in ICU
eFigure 1. Cumulative Blood Loss in the 48 Hours Post-Surgery
eFigure 2. Cumulative Blood Components Transfused in the 48 Hours Post-Surgery
eFigure 3. Time to Event Analysis: Days Until Commencement of Aspirin
eFigure 4. Time to Event Analysis: Days Until Commencement of Heparin
eFigure 5. Time to Event Analysis: Days to Extubation
eFigure 6. Time to Event Analysis: Days to ICU Discharge
eFigure 7. Time to Event Analysis: Days to Hospital Discharge
eFigure 8. Time to Event Analysis: Survival
eFigure 9. Subgroup Analysis of the Primary Outcome
eFigure 10. Cumulative proportion of Patients Receiving Early Open-Label Transfusion (Before 3 units), by Enrolment Order Within Site
Nonauthor Collaborators. Cryopreserved vs Liquid Platelets II (CLIP-II) Investigators
Data Sharing Statement
References
- 1.Rossaint R, Afshari A, Bouillon B, et al. The European guideline on management of major bleeding and coagulopathy following trauma: sixth edition. Crit Care. 2023;27(1):80. doi: 10.1186/s13054-023-04327-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mitra B, Jorgensen M, Reade MC, et al. ; Clinical and Consumer Reference Group for the Update of Patient Blood Management Guidelines (Module 1: Critical Bleeding/Massive Transfusion) . Patient blood management guideline for adults with critical bleeding. Med J Aust. 2024;220(4):211-216. doi: 10.5694/mja2.52212 [DOI] [PubMed] [Google Scholar]
- 3.Saluja K, Thakral B, Marwaha N, Sharma RR. Platelet audit: assessment and utilization of this precious resource from a tertiary care hospital. Asian J Transfus Sci. 2007;1(1):8-11. doi: 10.4103/0973-6247.28066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fedele PL, Polizzotto MN, Grigoriadis G, et al. Profiling clinical platelet and plasma use to inform blood supply and contingency planning: PUPPY, the prospective utilization of platelets and plasma study. Transfusion. 2016;56(10):2455-2465. doi: 10.1111/trf.13778 [DOI] [PubMed] [Google Scholar]
- 5.Young PP, Ehrhart J, Mair D. Rural hospitals and access to platelets for treating active bleeding: is the heartland bleeding? Transfusion. 2020;60(11):2474-2475. doi: 10.1111/trf.15970 [DOI] [PubMed] [Google Scholar]
- 6.Valeri CR. Hemostatic effectiveness of liquid-preserved and previously frozen human platelets. N Engl J Med. 1974;290(7):353-358. doi: 10.1056/NEJM197402142900702 [DOI] [PubMed] [Google Scholar]
- 7.Valeri CR, Ragno G, Khuri S. Freezing human platelets with 6 percent dimethyl sulfoxide with removal of the supernatant solution before freezing and storage at -80 degrees C without postthaw processing. Transfusion. 2005;45(12):1890-1898. doi: 10.1111/j.1537-2995.2005.00647.x [DOI] [PubMed] [Google Scholar]
- 8.Lelkens CC, Koning JG, de Kort B, Floot IB, Noorman F. Experiences with frozen blood products in the Netherlands military. Transfus Apher Sci. 2006;34(3):289-298. doi: 10.1016/j.transci.2005.11.008 [DOI] [PubMed] [Google Scholar]
- 9.Johnson L, Reade MC, Hyland RA, Tan S, Marks DC. In vitro comparison of cryopreserved and liquid platelets: potential clinical implications. Transfusion. 2015;55(4):838-847. doi: 10.1111/trf.12915 [DOI] [PubMed] [Google Scholar]
- 10.Slichter SJ, Dumont LJ, Cancelas JA, et al. Safety and efficacy of cryopreserved platelets in bleeding patients with thrombocytopenia. Transfusion. 2018;58(9):2129-2138. doi: 10.1111/trf.14780 [DOI] [PubMed] [Google Scholar]
- 11.Bohonek M, Kutac D, Landova L, et al. The use of cryopreserved platelets in the treatment of polytraumatic patients and patients with massive bleeding. Transfusion. 2019;59(S2):1474-1478. doi: 10.1111/trf.15177 [DOI] [PubMed] [Google Scholar]
- 12.Khuri SF, Healey N, MacGregor H, et al. Comparison of the effects of transfusions of cryopreserved and liquid-preserved platelets on hemostasis and blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg. 1999;117(1):172-183. doi: 10.1016/S0022-5223(99)70483-6 [DOI] [PubMed] [Google Scholar]
- 13.Reade MC, Marks DC, Bellomo R, et al. ; Cryopreserved vs Liquid Platelet (CLIP) Investigators, the Australian and New Zealand College of Anaesthetists Clinical Trials Network, and the Australian and New Zealand Intensive Care Society Clinical Trials Group . A randomized, controlled pilot clinical trial of cryopreserved platelets for perioperative surgical bleeding: the CLIP-I trial (editorial, p 2759). Transfusion. 2019;59(9):2794-2804. doi: 10.1111/trf.15423 [DOI] [PubMed] [Google Scholar]
- 14.McGuinness S, Charlewood R, Gilder E, et al. ; Australian and New Zealand College of Anaesthetists Clinical Trials Network; Australian and New Zealand Intensive Care Society Clinical Trials Group . A pilot randomized clinical trial of cryopreserved versus liquid-stored platelet transfusion for bleeding in cardiac surgery: the cryopreserved versus liquid platelet-New Zealand pilot trial. Vox Sang. 2022;117(3):337-345. doi: 10.1111/vox.13203 [DOI] [PubMed] [Google Scholar]
- 15.McQuilten ZK, Andrianopoulos N, Wood EM, et al. Transfusion practice varies widely in cardiac surgery: results from a national registry. J Thorac Cardiovasc Surg. 2014;147(5):1684-1690.e1. doi: 10.1016/j.jtcvs.2013.10.051 [DOI] [PubMed] [Google Scholar]
- 16.Flint AWJ, Bailey M, Reid CM, et al. Preoperative identification of cardiac surgery patients at risk of receiving a platelet transfusion: the Australian Cardiac Surgery Platelet Transfusion (ACSePT) risk prediction tool. Transfusion. 2020;60(10):2272-2283. doi: 10.1111/trf.15990 [DOI] [PubMed] [Google Scholar]
- 17.Reade MC, Marks DC, Howe B, et al. ; CLIP-II and CLIPNZ-II Investigators.; CLIP-II and CLIPNZ-II Investigators . Cryopreserved platelets compared with liquid-stored platelets for the treatment of surgical bleeding: protocol for two multicentre randomised controlled blinded non-inferiority trials (the CLIP-II and CLIPNZ-II trials). BMJ Open. 2022;12(12):e068933. doi: 10.1136/bmjopen-2022-068933 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lehmann EL. Nonparametrics: Statistical Methods Based on Ranks. Springer; 2006. [Google Scholar]
- 19.Blood component information (BCI): an extension of blood component labels, 2025 edition. Australian Red Cross Lifeblood; 2024. Accessed November 17, 2025. https://www.lifeblood.com.au/sites/default/files/resource-library/2025-09/J25247_BCI_0.pdf
- 20.Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41(4):734-744. doi: 10.1093/ejcts/ezs043 [DOI] [PubMed] [Google Scholar]
- 21.Madhavan S, Chan SP, Tan WC, et al. Cardiopulmonary bypass time: every minute counts. J Cardiovasc Surg (Torino). 2018;59(2):274-281. [DOI] [PubMed] [Google Scholar]
- 22.Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation. 2011;123(23):2736-2747. doi: 10.1161/CIRCULATIONAHA.110.009449 [DOI] [PubMed] [Google Scholar]
- 23.Ranieri VM, Rubenfeld GD, Thompson BT, et al. ; ARDS Definition Task Force . Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307(23):2526-2533. [DOI] [PubMed] [Google Scholar]
- 24.Waters L, Padula MP, Marks DC, Johnson L. Cryopreserved platelets demonstrate reduced activation responses and impaired signaling after agonist stimulation. Transfusion. 2017;57(12):2845-2857. doi: 10.1111/trf.14310 [DOI] [PubMed] [Google Scholar]
- 25.Alghamdi AA, Davis A, Brister S, Corey P, Logan A. Development and validation of Transfusion Risk Understanding Scoring Tool (TRUST) to stratify cardiac surgery patients according to their blood transfusion needs. Transfusion. 2006;46(7):1120-1129. doi: 10.1111/j.1537-2995.2006.00860.x [DOI] [PubMed] [Google Scholar]
- 26.Rothermel LD, Lipman JM. Estimation of blood loss is inaccurate and unreliable. Surgery. 2016;160(4):946-953. doi: 10.1016/j.surg.2016.06.006 [DOI] [PubMed] [Google Scholar]
- 27.Saris A, Pavenski K. Human leukocyte antigen alloimmunization and alloimmune platelet refractoriness. Transfus Med Rev. 2020;34(4):250-257. doi: 10.1016/j.tmrv.2020.09.010 [DOI] [PubMed] [Google Scholar]
- 28.Holley A, Marks DC, Johnson L, Reade MC, Badloe JF, Noorman F. Frozen blood products: clinically effective and potentially ideal for remote Australia. Anaesth Intensive Care. 2013;41(1):10-19. doi: 10.1177/0310057X1304100104 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
Investigator Brochure
Trial Supplementary Information
eTable 1. Additional Baseline Data
eTable 2. Study Hospitals
eTable 3. Predefined Tertiary and Additional Effectiveness Outcomes
eTable 4. Viscoelastic Testing Results After the Last Study Platelet Transfusion
eTable 5. Additional Safety Outcome Data
eTable 6. Baseline Characteristics of Patients in the Sensitivity Analysis (Excluding Recipients of Open Label Prior to 3 Study Units)
eTable 7. Characteristics of Included vs Excluded Patients (Received Open Label Prior to 3 Study Units)
eTable 8. Primary and Secondary Outcomes for The Sensitivity Analysis (Excluding Patients That Received Open Label Prior to 3 Study Units (n=29))
eTable 9. Mixed-Effects Model Estimates of Lowest Platelet Count in the First 24 Hours in ICU
eFigure 1. Cumulative Blood Loss in the 48 Hours Post-Surgery
eFigure 2. Cumulative Blood Components Transfused in the 48 Hours Post-Surgery
eFigure 3. Time to Event Analysis: Days Until Commencement of Aspirin
eFigure 4. Time to Event Analysis: Days Until Commencement of Heparin
eFigure 5. Time to Event Analysis: Days to Extubation
eFigure 6. Time to Event Analysis: Days to ICU Discharge
eFigure 7. Time to Event Analysis: Days to Hospital Discharge
eFigure 8. Time to Event Analysis: Survival
eFigure 9. Subgroup Analysis of the Primary Outcome
eFigure 10. Cumulative proportion of Patients Receiving Early Open-Label Transfusion (Before 3 units), by Enrolment Order Within Site
Nonauthor Collaborators. Cryopreserved vs Liquid Platelets II (CLIP-II) Investigators
Data Sharing Statement
