TiME OUT: Time-specific machine-learning evaluation to optimize ultramassive transfusion.

BACKGROUND: Ultramassive transfusion (UMT) is a resource-demanding intervention for trauma patients in hemorrhagic shock, and associated mortality rates remains high. Current research has been unable to identify a transfusion ceiling or point where UMT transitions from lifesaving to futility. Furthermore, little consideration has been given to how time-specific patient data points impact decisions with ongoing high-volume resuscitation. Therefore, this study sought to use time-specific machine learning modeling to predict mortality and identify parameters associated with survivability in trauma patients undergoing UMT. METHODS: A retrospective review was conducted at a Level I trauma (2018-2021) and included trauma patients meeting criteria for UMT, defined as ≥20 red blood cell products within 24 hours of admission. Cross-sectional data were obtained from the blood bank and trauma registries, and time-specific data were obtained from the electronic medical record. Time-specific decision-tree models predicating mortality were generated and evaluated using area under the curve. RESULTS: In the 180 patients included, mortality rate was 40.5% at 48 hours and 52.2% overall. The deceased received significantly more blood products with a median of 71.5 total units compared with 55.5 in the survivors ( p < 0.001) and significantly greater rates of packed red blood cells and fresh frozen plasma at each time interval. Time-specific decision-tree models predicted mortality with an accuracy as high as 81%. In the early time intervals, hemodynamic stability, undergoing an emergency department thoracotomy, and injury severity were most predictive of survival, while, in the later intervals, markers of adequate resuscitation such as arterial pH and lactate level became more prominent. CONCLUSION: This study supports that the decision of "when to stop" in UMT resuscitation is not based exclusively on the number of units transfused but rather the complex integration of patient and time-specific data. Machine learning is an effective tool to investigate this concept, and further research is needed to refine and validate these time-specific decision-tree models. LEVEL OF EVIDENCE: Prognostic and Epidemiological; Level IV.

Anti-Rh alloimmunization after trauma resuscitation.

The AABB recently posted a bulletin (19-02) regarding their recommendations for the use of group O red blood cells (RBCs) during trauma. Though group O Rh(D)-negative RBC units are considered the 'safest', the demand of such units often exceeds the supply. Therefore, O Rh(D)-positive units are often used during the first parts of a massive transfusion protocol (MTP) or patients with particularly severe hemorrhage are switched over from O Rh(D)-negative to O Rh(D)-positive RBC units in order to preserve the O Rh(D)-negative supply. In light of these limitations, it is important to understand the risk of such policies to the patient. The reported risk of alloimmunization after exposure to Rh(D)-positive RBCs ranges widely from 3 to 70%. In response, we performed a retrospective review of 1,198 patients in our institution that had a MTP activation due to trauma. Of those patients, we focused on Rh(D)-negative patients that received at least 1 unit of Rh(D)-positive RBCs. Seventy-two patients met the criteria for inclusion, accounting for 6% of the total population. Of the 72 Rh(D)-negative patients, we identified 17% that formed new Rh group antibodies after exposure to Rh(D)-positive RBCS. All 10 of our alloimmunized patients (two of which were females of childbearing age) formed anti-D, while 3 patients also formed either anti-E or anti-C. Since this was a retrospective review, we did not perform repeated antibody screens for the entire study period, but did review all records for the entire period. We did note that we were more likely to detect an novel alloantibody if more antibody screens were performed during the patient's initial stay and during follow-up visits. We conclude that providing Rh(D) negative patients Rh(D) positive RBC units is not without risk and policies regarding such provisions should be carefully considered. As RBC shortages continue to be a part of daily practice, such issues may continue to be a challenge for the blood bank community.

Validation and cost-effectiveness of an in-house dithiothreitol (DTT) treatment protocol for daratumumab patients in a large tertiary care hospital provides gateway for implementation in smaller community hospitals.

BACKGROUND: Treatment of multiple myeloma with daratumumab (DARA) is increasing fast. Unfortunately, this antibody also attaches to red blood cells (RBCs) and mimics an autoantibody's panreactivity during pre-transfusion testing, necessitating specialized techniques, (e.g. dithiothreitol (DTT)) for alloantibody detection. Many hospitals use a reference lab for such testing, increasing both cost and turn-around time (TAT). Herein, we compare the cost and TAT, pre and post-implementation of an in-house DTT protocol. METHODS: We designed a validation of our in-house DTT protocol from Nov to Dec 2017 with full implementation on January 1, 2018. We retrospectively reviewed all pre-transfusion tests on DARA patients from Feb 2016 to April 2018, pre and post-implementation of in-house DTT testing. Descriptive statistics were used for patient demographics and a Student t-test was used to compare cost and TATs (pre and post-implementation). RESULTS: We identified 49 patients on DARA treatment requiring transfusion. Samples from these patients were sent to the reference lab 104 times and were tested in-house 28 times. The average TAT for the reference lab was 19h25 m compared to our in-house TAT of 5h9m (an average time-savings of 14h16 m). We spent approximately $33,800 ($325 per test) for 104 reference lab samples versus $806.12 (˜$28.79 per test) for in-house testing of 28 samples. CONCLUSION: We provide an easily implementable DTT protocol for pre-transfusion testing community hospitals and beyond. As more monoclonal antibodies are developed and approved for clinical use, the lessons learned with DARA will expand to deal with interference from future targeted therapies.

Solving the calcium gluconate shortage in real-time: Mistakes made and lessons learned.

PURPOSE: During a national shortage of calcium gluconate, we switched to calcium chloride for routine supplementation for peripheral blood stem cell (PBSC) collections. Subsequently, we analyzed the postprocedure ionized calcium level, as we aimed for an equivalent result compared to before the shortage. METHODS: Pharmacy representatives helped us to find an "equivalent" substitute for calcium gluconate at 46.5 mEq in 500 mL normal saline, infused at 100 mL/hour. After instituting a presumably comparable protocol using calcium chloride (40.8 mEq in 250 mL normal saline at a rate of 100 mL/hour), we reviewed ionized calcium results post-PBSC procedures to compare with those obtained with calcium gluconate. Having noticed a difference in the mean values, we adjusted the rate of calcium chloride to reach our desired outcome. RESULTS: Twenty-seven procedures were analyzed on 15 unique patients. We used the Spectra OPTIA with a whole blood: anticoagulant ratio of 13:1. Ionized calcium levels post-PBSC collection with the first calcium chloride protocol were significantly higher (P = 0.003) in nine patients treated. Subsequently, we decreased the calcium chloride infusion rate to 75 mL/hour and achieved similar mean levels to calcium gluconate (P = 0.382). CONCLUSION: Changes in replacement fluids for apheresis procedures can be complex, particularly when dealing with electrolytes that could be clinically significant at critically high or low levels. Once we recognized the need to take into account the amount of elemental calcium infused, we achieved the desired postprocedure ionized calcium results. This study can serve as a lesson for future shortages of infusions used during apheresis procedures.

Packed red blood cell donor age affects overall survival in transfused patients undergoing hepatectomy for non-hepatocellular malignancy.

BACKGROUND: Patients undergoing hepatectomy often require packed red blood cell (PRBC) transfusion, which has been associated with worse oncologic outcomes. However, limited data exist regarding the impact of PRBC donor factors. We hypothesized that PRBC donor age impacts survival after hepatectomy for non-hepatocellular malignancies. METHODS: Patients who underwent hepatectomy for non-hepatocellular malignancy from 2005 to 2014 were retrospectively evaluated. Impact of clinicopathologic and PRBC factors on oncologic outcomes were assessed. RESULTS: Of 149 identified patients, 76 received a perioperative PRBC transfusion (median 2 units). Transfusion was associated with increased median length of stay (8 vs. 6 days; p < 0.01) and median operative blood loss (700 vs. 350 mL; p < 0.01) versus non-transfused, respectively. In transfused patients, receipt of PRBC from older donors compared to younger resulted in decreased RFS (0.94 vs. 2.63 years, respectively; p = 0.02) and OS (1.94 vs. 3.44 years, respectively; p = 0.6). The PRBC donor age was an independent predictor of decreased recurrence free survival on multivariate analysis (HR 2.5, p = 0.04). CONCLUSIONS: In patients undergoing hepatectomy for non-hepatocellular malignancies and receiving perioperative transfusion, PRBC donor age may impact survival and warrants further investigation.

Soluble antigens in plasma allow mismatched transfusion without hemolysis.

BACKGROUND: Universal plasma is a scarce resource when a massive transfusion protocol has been initiated. Previous studies have reported success using group A plasma in place of the universal plasma, group AB. It is unclear why there are not more reports of hemolytic reactions occurring from this practice. One possible explanation is the presence of water-soluble antigens in the patient plasma that bind to, and neutralize, the soluble antibodies present in the transfused plasma. STUDY DESIGN AND METHODS: Expired units of plasma were used to make dilutions that consisted of mixtures of group A and B plasma and saline. Serial dilutions of these samples were performed starting from undiluted up to 1024. The dilutions were titrated using a group B red blood cell preparation. The titrations were read after incubation. RESULTS: The titers that resulted from the mixed plasma dilutions were significantly lower or showed no agglutination when compared to the group A-specific saline dilutions. The differences between the saline dilutions and mixed group dilutions were significant (p < 0.001). CONCLUSION: Our study shows that secretor status would provide protection from isoantibodies. The dissolved B antigens in the group B plasma absorb and/or bind to the group B isoantibodies in the group A plasma. This mechanism gives a protective effect against hemolytic reactions in massive transfusion situations in the trauma setting when group A plasma is used instead of group AB plasma. This protective effect is revealed with the paucity of intravascular hemolysis observed in these out-of-group massive transfusions.

Analysis of responses to the Rho-kinase inhibitor Y-27632 in the pulmonary and systemic vascular bed of the rat.

Responses to the Rho kinase inhibitor Y-27632 were investigated in the anesthetized rat. Under baseline conditions intravenous injections of Y-27632 decreased pulmonary and systemic arterial pressures and increased cardiac output. The decreases in pulmonary arterial pressures were enhanced when baseline tone was increased with U-46619, and under elevated tone conditions Y-27632 produced similar percent decreases in pulmonary and systemic arterial pressures. Injections of Y-27632 prevented and reversed the hypoxic pulmonary vasoconstrictor response. The increase in pulmonary arterial pressure in response to ventilation with a 10% O(2)-90% N(2) gas mixture was not well maintained during the period of hypoxic exposure. Treatment with the nitric oxide (NO) synthase (NOS) inhibitor nitro-l-arginine methyl ester (l-NAME) increased pulmonary arterial pressure and prevented the decline or fade in the hypoxic pulmonary vasoconstrictor response. The hypoxic pulmonary vasoconstrictor response was reversed by Y-27632 in control and in l-NAME-treated animals. The Rho kinase inhibitor attenuated increases in pulmonary arterial pressures in response to intravenous injections of serotonin, angiotensin II, and Bay K 8644. Y-27632, sodium nitrite, and BAY 41-8543, a guanylate cyclase stimulator, decreased pulmonary and systemic arterial pressures and vascular resistances in monocrotaline-treated rats. These data suggest that Rho kinase is involved in the regulation of baseline tone and in the mediation of pulmonary vasoconstrictor responses. The present data suggest that the hypoxic pulmonary vasoconstrictor response is modulated by the release of NO that mediates the nonsustained component of the response in the anesthetized rat. These data suggest that Rho kinase and NOS play important roles in the regulation of vasoconstrictor tone in physiological and pathophysiological states and that monocrotaline-induced pulmonary hypertension can be reversed by agents that inhibit Rho kinase, generate NO, or stimulate soluble guanylate cyclase.