In the present study, we evaluated the survival benefit of cryoprecipitate use in the resuscitation phase of trauma patients requiring massive transfusion, in which the amounts of administered clotting factors were considered. The results suggested that the concurrent use of cryoprecipitate and FFP is more beneficial than the administration of FFP alone when supplementing the same amounts of clotting factors. Although significantly higher mortality was observed in the Cryo + FFP group in the overall cohort (not the propensity score-matched cohort), cryoprecipitate is generally used for more severe patients requiring a larger amount of transfusion, as shown in Table 1. After the appropriate case-mix adjustment, the potential advantage of early administration of cryoprecipitate was observed. From the cumulative incidence curve, the difference in mortality was large in the first 1 to 2 days and then maintained, suggesting that the hemostatic effect mainly caused the observed benefit.
The concentration ratios of clotting proteins in cryoprecipitate, compared to FFP, varied according to the type of clotting factor. For example, the concentrations of fibrinogen, factor VIII, and VWF have been estimated to be approximately 3, 8, and 12 times that of FFP, respectively, when cryoprecipitate is processed through blood banks in the United States. However, the concentration of clotting factors varies depending, to some extent, on the blood donors [16]. In the present study, we assumed a uniform conversion ratio between cryoprecipitate and FFP; however, the non-uniformity of concentration ratios in each cryoprecipitate product and the effects of clotting factors other than fibrinogen should also be considered. Although we also explored differences in outcomes according to different conversion ratios, assumptions based on a conversion using a certain ratio would not be the optimal approach. However, since a large-scale study evaluating the effectiveness of cryoprecipitate considering the amount of administered clotting factors has not been reported, the present study results provide significant clinical evidence on this aspect before large-scale randomized controlled trials are conducted.
Although the granular mechanism of cryoprecipitate could not be discussed from this registry-based analysis, there were several possible reasons for the significantly lower mortality rate observed in patients who received cryoprecipitate. If clotting factors are administered only with the purpose of supplementation in patients sustaining hemorrhagic shock, transfusion of whole blood or blood components with a 1:1:1 ratio of pRBC:FFP:platelets would suffice. However, in cases of severe injury, the presence of injury-induced coagulopathy, in addition to hemorrhage and blood dilution due to fluid resuscitation, is also common [29, 30]. The results of this study have highlighted the benefits of cryoprecipitate administered as a treatment against trauma-induced coagulopathy, which accompanies the dysregulated fibrinolytic system and a relative deficiency in clotting factors observed in these patients. Furthermore, significant fluid overload is inevitable when administering large amounts of clotting factors by infusion of FFP only, which has been reported to have a negative impact on trauma patient outcomes [31–33]. Although the information on the administered fluid volume was not available in the TQIP database, appropriate volume management might have been achieved by administering concentrated clotting factors using cryoprecipitate. The RETIC trial [34] showed that the need for rescue therapies significantly decreased in patients who received clotting factor concentrates, primarily fibrinogen concentrate, compared to patients who received FFP only. We also evaluated the incidence of complications in the present study; however, no significant differences were observed between groups. Although a survivor bias was unavoidable in assessing this outcome measure, it works in an unfavorable direction in patients who survive longer, suggesting that cryoprecipitate could improve patient survival without increasing complications. The immunomodulatory effects of plasma also might have affected patient outcomes. Plasma and cryoprecipitate made from plasma contain various important proteins such as endothelial glycocalyx, fibronectin, and platelet microparticles, in addition to clotting factors, which have important biologic functions and are said to regulate vascular permeability [35, 36]. Although the concentrations of these immunomodulatory proteins in cryoprecipitate have not been well-studied, potential differences in the amount of administered plasma proteins between cryoprecipitate and FFP could be a possible explanation for the observed differences in the outcomes.
Clinical evidence for the comparative effectiveness of cryoprecipitate and other clotting factor preparations has been limited in bleeding patients [37, 38]. A theoretical advantage of cryoprecipitate use is that it contains a variety of important coagulation factors. It has been reported that the levels of VWF and factor XIII, which play critical roles in primary and secondary hemostasis, are significantly decreased in trauma patients who experience major bleeding [39]. Moreover, fibrinogen concentrate manufacturing costs are, on average, 3 to 4 times higher than that of cryoprecipitate preparation [40]. However, there have also been concerns raised relative to the risk of viral infection using cryoprecipitate preparations. As opposed to fibrinogen concentrate, which is virally inactivated, the safety of cryoprecipitate use is still of concern, although several efforts for pathogen reduction have been made [12]. Furthermore, cryoprecipitate requires a thawing time of approximately 17 to 20 min, in contrast to fibrinogen concentrate which can be stored in the ED at room temperature, allowing prompt administration when needed. Further clinical evidence is required to establish a next-stage optimal management strategy against trauma-induced coagulopathy in patients who require massive transfusions.
Strength and Limitations
This study had several strengths. First, we analyzed the large-scale trauma registry collecting real-world data of North America in which many patients who received massive transfusions were included. The amount of administered transfusion components and cryoprecipitate within the first 4 h were recorded in the database, which enabled us, for the first time, to evaluate the comparative benefits of cryoprecipitate in addition to FFP in hemostatic resuscitation. However, several limitations must be acknowledged. The issue of residual confounding was inevitable due to the retrospective nature of the study. Information on the laboratory test results, details about the time-course in the ED, use of tranexamic acid, and detailed hemostatic procedures were not available in the TQIP database. Also, the information on the administered fluid volume was unavailable in the TQIP database, which prevented assessing the dilutional effect. The use of fibrinogen concentrate could not be identified because the procedure code was not defined in the International Classification of Diseases 10th Revision before 2016, although a small amount of clotting factors in blood products was accounted for by matching administered pRBCs and platelets volume. The guideline for massive transfusion by the American College of Surgeons suggested using cryoprecipitate according to viscoelastic coagulation measures [41]; however, information on the use of the test was also unavailable. Although many patients were evaluated, generalizability was limited because the TQIP is a trauma database in the USA, exckuding patients younger than 16 years and older than 90 years. The accuracy and missing data in the database could also be considered a limitation. As discussed above, making assumptions about the conversion ratio between cryoprecipitate and FFP at a certain level would not be an optimal approach. Finally, the hemostatic resuscitation strategy was not uniform across trauma centers; each center treated patients with different strategies without consistency. Hospitals with well-established transfusion protocols, including the use of cryoprecipitate, might have achieved better outcomes. Due to the lack of information on the unique identifier of a treating hospital, the model considering the hospital-level clustering could not be used in our analysis.