Related Experiment Video
Updated: Oct 1, 2025
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Cryoprecipitate use during massive transfusion: A propensity score analysis
Andrew M Fleming1, Kinjal S Shah2, Saskya E Byerly3
1Department of Surgery, University of Tennessee Health Science Center, 910 Madison Ave, 2nd Floor, Memphis, TN, United States.
Introduction:
Cryoprecipitate is frequently administered as an adjunct to balanced transfusion in the setting of traumatic hemorrhage. However, civilian studies have not demonstrated a clear survival advantage, and prior observational studies noted selection bias when analyzing cryoprecipitate use. Additionally, due to the logistics involved in cryoprecipitate administration, it is inconsistently implemented alongside standardized massive transfusion protocols. This study aims to evaluate the effects of early cryoprecipitate administration on inpatient mortality in the setting of massive transfusion for exsanguinating trauma and to use propensity score analysis to minimize selection bias.
Methods:
The registry of an urban level 1 trauma center was queried for adult patients who received at least 6 units of packed red blood cells within 4 h of presentation. Univariate analysis, multiple logistic regression, and propensity score matching were performed.
Results:
562 patients were identified. Patients with lower median RTS (6.86 (IQR 4.09-7.84) vs 7.6 (IQR 5.97-7.84), P<0.01), decreased Glasgow coma scale (12 (IQR 4-15) vs 15 (IQR 10-15), P<0.01), and increased lactate (7.5 (IQR 4.3-10.2) vs 4.9 (IQR 3.1-7.2), P<0.01) were more commonly administered cryoprecipitate. Mortality was greater among those who received cryoprecipitate (40.2% vs 23.7%, p<0.01) on univariate analysis. Neither multiple logistic regression (OR 0.917; 95% confidence interval 0.462-1.822; p = 0.805) nor propensity score matching (average treatment effect on the treated 2.3%, p = 0.77) revealed that cryoprecipitate administration was associated with a difference in inpatient mortality.
Conclusions:
Patients receiving cryoprecipitate within 4 h of presentation were more severely injured at presentation and had increased inpatient mortality. Multivariable logistic regression and propensity score analysis failed to show that early administration of cryoprecipitate was associated with survival benefit for exsanguinating trauma patients. The prospect of definitively assessing the utility of cryoprecipitate in exsanguinating hemorrhage warrants prospective investigation.
