Clinical Features and Risk Factors Analysis for Hemorrhage in Adults on ECMO

Background: The use of extracorporeal membrane oxygenation (ECMO) to support critically ill patients with cardiorespiratory dysfunction has increased over the last decades. However, hemorrhagic complications occur frequently during ECMO support, and this has a significant impact on morbidity and mortality. Thus, this study aimed to identify the risk factors for hemorrhage in patients receiving ECMO.Methods: Our retrospective study included 60 patients, who were admitted to the Taihe Hospital in Shiyan City, Hubei Province, China from February 2017 to October 2020. About 18 patients developed hemorrhagic complications, and 42 patients did not demonstrate such complications. Data regarding patient demography, laboratory tests, and clinical manifestations prior to ECMO were collected to analyze their clinical features. Univariable and multivariable logistic analyses were used to explore the risk factors for hemorrhage in adults on ECMO. The receiver operating characteristic (ROC) curve was used to evaluate the predictive value of the binary logistic model. The amount of blood transfusions was compared between the two groups, and the activated partial thromboplastin time (APTT), platelet count, and hemoglobin level before the initiation of ECMO.Results: Logistic analysis showed that a longer duration of ECMO support, higher APTT, and lower platelet count prior to ECMO were independent risk factors for hemorrhage in adults on ECMO. In addition, we found that the cannula site was the most common bleeding site. Most bleeding events occurred within the first 3 days of ECMO therapy. After the ECMO initiation, APTT was prolonged while the platelet count and hemoglobin levels were decreased. The amount of blood transfusion was significantly higher in the hemorrhage group than in the non-hemorrhage group.Conclusions: Clinicians should evaluate the risk of hemorrhage based on the coagulation function of patients, underlying disease, and the duration of ECMO support. In the first 3 days during ECMO support, special attention should be given to the cannula site, mucosal, and dermal regions, and digestive tract to detect any signs of hemorrhage. Moreover, increasing the platelet count transfusion threshold and accurately determining the amount of blood transfusion required may prevent bleeding events.

O-EGS04 Is group and save necessary for all patients undergoing emergency cholecystectomy: A 6-year retrospective audit

Background Emergency cholecystectomy (EC) has a low perioperative bleeding risk. There is no current national guideline to suggest routine preoperative Group and Save (G&S) is necessary. Our Trust guideline recommends preoperative G&S for all EC operations. In 2018, a Trust-wide policy was adopted based on an audit, which concluded that routine preoperative G&S is unnecessary for elective cholecystectomy. All G&S require 2samples taken separately, which can delay surgery. The cost to process one sample for G&S is £28. Therefore, a study was set up to assess the need for routine G&S in patients undergoing EC. Methods This retrospective observational study was based on a prospectively collected hospital database from March 2015 to March 2021 using MS-Excel. All patients who underwent EC (laparoscopic and/or open) within 10 days during index admission were included. All elective cholecystectomies were excluded. Patients were divided into GS-patients (patients with G&S) and NGS-patients (patients without G&S). The primary outcome is the difference between the incidence of ‘Perioperative blood transfusion’ (PBT) between the studied groups. The overall cost-effectiveness is considered as a secondary outcome. The categorical data were analysed using the Chi-square test; a p-value <0.05 is considered statistically significant. Results In this 6year period, 2210patients underwent cholecystectomy. Of these, 496patients (22.4%) who underwent EC were included. 447patients (90.1%) were in GS group and 49patients (9.9%) were in the NGS group. None from the NGS group required PBT, whereas 3patients (0.6%) in the GS group received blood transfusion. However, PBT was truly indicated in 1patient due to the associated cardiovascular comorbidities. On the contrary, 2patients did not fit the ‘restrictive transfusion threshold’ criteria of JPAC. There was no statistically significant difference in PBT requirement between the studied groups (p = 0.331). Deferring routine G&S for EC could have saved our institution £24,976. Conclusions Our study has demonstrated that preoperative G&S is not indicated for all emergency cholecystectomies. It takes approximately 1 hour for G&S to be processed unless crossmatching is required. Group O-negative or O-positive blood can be provided to patients when urgent blood transfusion is needed depending on their age and gender. Thus, we conclude that G&S should be restricted to patients with low preoperative haemoglobin, bleeding and clotting disorders, those known to have abnormal blood antibodies and significant cardiovascular comorbidities.

Abstract 9115: Determining Optimal Hemoglobin Transfusion Threshold for Patients With Severe Traumatic Brain Injury: A Systematic Review and Meta-Analysis

Introduction: Multiple randomized controlled trials have found that a conservative approach to transfusing critically ill patients reduces mortality, with current guidelines recommending a hemoglobin (HgB) transfusion threshold of 7 g/dL. However, little work has examined whether this transfusion threshold reduces mortality in patients with severe traumatic brain injury (TBI). Here, we present a systematic review and meta-analysis of the literature. Methods: A systematic search was conducted on PubMed, Ovid, and Web of Science. Full-text articles were eligible if patients with TBI, defined as Glasgow Coma Score <= 8, were divided into multiple groups with varying hemoglobin transfusion thresholds and reported any outcome of interest including mortality, number of packed red blood cell (PRBC) units transfused, length of stay in ICU, and length of stay in the hospital. Eight studies were eligible (n = 3663). We compared mortality rates at HgB transfusion thresholds of < 7 g/dL, < 8 g/dL, < 9 g/dL, and < 10 g/dL. Results: We found that traditionally ‘conservative’ approaches to anemia management (HgB < 7 g/dL, < 8 g/dL, and < 9 g/dL) were associated with decreased mortality when compared to traditionally ‘liberal’ approaches (HgB < 10 g/dL), with p < 0.05. Results were robust across both frequentist and Bayesian analysis. As a surrogate for cost of care and use of hospital resources, the total number of PRBC units transfused to patients, length of stay in ICU, and length of stay in hospital were analyzed. We found that using a transfusion threshold < 7 g/dL compared to < 10 g/dL substantially decreased the number of PRBC units transfused. In three of five cohorts, the cohort with the lower HgB transfusion threshold or no transfusion had a significantly shorter length of stay in the ICU and in the hospital. The remaining two cohorts found no significant difference in the length of stays in ICU or hospital. Conclusion: This study demonstrates that conservative approaches to transfusions ( < 7 g/dL, < 8 g/dL, or < 9 g/dL) significantly reduce mortality and the number of PRBC units transfused when compared to more liberal approaches ( < 10 g/dL). Current evidence is unclear on the benefits of conservative approaches in reduction of ICU or hospital length of stay.

Platelet Transfusions Prior to Lumbar Punctures, What Is the Evidence?

Introduction: Lumbar punctures (LPs) are routinely used to administer intrathecal chemotherapy for children and adults with hematologic malignancies. The platelet threshold at which an LP can be safely performed is a debated topic. A platelet count of ≥50x10 9/L is a widely accepted threshold despite sparse evidence supporting this number (Kaufman et al 2003). Previous studies failed to report any increase in the incidence of post-procedure adverse events in patients with a platelet count below this threshold who undergo LPs (Corroa et al 2018, Chung et al 2020, Öztürk et al 2021). However, these prior studies were limited in their sample size for the number of LPs done in patients with platelets &lt;50 x 10 9/L. Objectives: To assess the effects of differing platelet transfusion thresholds (&lt;50 x 10 9/L versus ≥50x10 9/L) in patients undergoing lumbar punctures. Methods: We performed a computerized search of Medline, Cochrane and Google Scholar databases through January 2021 for studies evaluating lumbar punctures in adult and pediatric patients. We searched for randomized control trials, non-randomized controlled trials, cohort studies, and retrospective case series. We included any study that involved transfusions of platelet concentrates and was given to prevent bleeding in people of any age with thrombocytopenia that required insertion of a lumbar puncture needle or epidural catheter. We screened 10502 studies and 17 of the studies met criteria for review and were included in the final review. Standard methodological review was completed for each study. Two review authors independently assessed studies for eligibility and extracted data. Results are expressed narratively, and a meta-analysis was performed evaluating traumatic taps with a platelet threshold of &lt;50 x10 9/Lcompared to those over the transfusion threshold of ≥50x10 9/L. Results: We identified 7 retrospective cohort studies and 10 case series that contained patients who received and did not receive any platelets prior to lumbar puncture procedures (Table 1). Most of the studies (12/17) had a patient population that included an oncologic diagnosis with hematologic malignancies being the most common (Table 1). We summarized the studies in narrative form with relevant outcomes (Table 1). Spinal hematoma as an outcome was documented in 12/17 studies. There was no difference in spinal hematomas in patients with platelet counts of &lt;50 x10 9/Lcompared to those over the transfusion threshold of ≥50x10 9/L (Table 1). We performed a meta-analysis in patients with platelet counts of &lt;50 x10 9/Lcompared to those over the transfusion threshold of ≥50x10 9/L for the outcome of traumatic taps. When traumatic taps were defined as greater than 10 x 10 6 RBCs/L, the incidence of traumatic taps favored the patients who had a platelet count ≥50x10 9/L (Figure 1). However, when traumatic taps were defined as &gt;400 RBCs/L the incidence of traumatic taps favored the patients who had a platelet count &lt;50 x10 9/L (Figure 1). There were no studies that evaluated all-cause mortality within 30 days of the procedure, length of stay, and quality of life outcomes in patients. Conclusion: We found no clinical trials to base our recommendations on platelet thresholds prior to a lumbar puncture. We did find multiple studies that should have enough patients to determine if a platelet threshold less than &lt;50 x10 9/L is associated with increased spinal hematomas. There was no increase in spinal hematomas when the platelet count &lt;50 x10 9/L was compared to a cohort with a platelet count ≥50x10 9/L. The meta-analysis of traumatic taps depended on the definition of traumatic taps and when traumatic taps were defined as greater than 10 x 10 6 RBCs/L, the incidence of traumatic taps favored the patients who had a platelet count ≥50x10 9/L. Whereas, when traumatic taps were defined as &gt;400 RBCs/L the incidence of traumatic taps favored the patients who had a platelet count &lt;50 x10 9/L. A multicenter prospective study is needed to clarify the incidence of traumatic taps and to verify the data regarding spinal hematomas, other adverse events or quality of life measures in patients. Figure 1 Figure 1. Disclosures Laber: Alexion: Membership on an entity's Board of Directors or advisory committees. Visweshwar: Biogen Idec: Membership on an entity's Board of Directors or advisory committees. Jaglal: Novartis: Membership on an entity's Board of Directors or advisory committees; Sanofi: Membership on an entity's Board of Directors or advisory committees.

Retrospective Cohort Study of Neonatal Blood Transfusion in China

Background: Blood transfusion treatment is extremely important for newborns，but the threshold of neonatal blood transfusion is not same in different countries, which may be due to differences in regions, races and nationalities, as well as medical conditions and treatment methods. Up to now, there are not enough clinical studies and prospective follow-up to determine the suitable threshold for Chinese newborns. Therefore, it is important to establish a retrospective and prospective multicenter cohort study to evaluate whether the blood transfusion scheme is suitable for newborns in China.Methods: This is a retrospective cohort study of neonatal blood transfusion and prospective follow-up from January 1, 2017 to June 30, 2021, aim to evaluate the effect of restricted and unrestricted blood transfusion on neonatal health. Diagnosis and blood transfusion data of 5,669 newborns between January 1, 2017 and June 30, 2018 from 46 hospitals in China were analyzed through retrospective study and followed up for 1w,1m and 3y after discharge. The variable data of newborns and their mothers was collected in this cohort study with 280 variables and 2.98 million data volumes including in the database. The primary outcome index of the study was death, and the secondary outcome index was complications during hospitalization, hospitalization time, NICU hospitalization days and hospitalization expenses.Discussion: The groups were grouped by birth weight, and each group was defined as a restricted and unrestricted cohort according to the Recommended Program for Neonatal Blood Transfusion (5th Edition), and evaluate applicability of this scheme for Chinese newborns based on outcome indicators. According to the neonatal treatment data, a appropriate neonatal blood transfusion threshold and neonatal blood transfusion program for China would be determined.

The Association Between Platelet Transfusion and Mortality Rate Among Preterm Neonates in the Eastern Province, Saudi Arabia

Platelet transfusion is the main mode of management of thrombocytopenia. However, some studies link frequent and high-threshold platelet transfusions with an incremental increase of mortality rate. Aim: This study aims to assess the association between the frequency and the threshold of platelet transfusions, with the mortality rate among preterm neonates. Methods: A retrospective cohort study design was used. This study was conducted at maternity and children's hospitals in Al-Ahasa, Saudi Arabia. The sample size includes 154 preterm neonates, included in the study by the use of the convenience sampling technique. Result: There is a significant relationship found between the gestational age and the birth weight of the preterm neonates with the survival rates among both groups. In contrast, there is no significant relationship found between transfusion frequency, transfusion threshold, and the survival rates of the group which received platelet transfusion. Implications for Practice and Research: The current study found that mortality is mainly associated with lower gestational ages, and not platelet transfusions. More studies are needed to fill the remaining gaps of knowledge, and to optimise platelet transfusion practices among preterm neonates. Key Words: platelet transfusion, preterm neonate, platelet transfusion threshold, mortality rate.

Seeking Strategies to Optimize Blood Utilization: The Preliminary Experience with Implementing a Patient Blood Management Program in a Greek Tertiary Hospital

Objectives: Our aim was to assess blood utilization after implementation of a patient blood management (PBM) program in a Greek tertiary hospital. Methods: An electronic transfusion request form and a prospective audit of transfusion practice were implemented. After the one-year implementation period, a retrospective review was performed to assess transfusion practice in medical patients. Results: Pre-PBM, a total of 9478 RBC units were transfused (mean: 1.75 units per patient) compared with 9289 transfused units (mean: 1.57 units per patient) post-PBM. Regarding the post-PBM period, the mean hemoglobin (Hb) level of the 3099 medical patients without comorbidities transfused was 7.19 ± 0.79 gr/dL. Among them, 2065 (66.6%) had Hb levels >7.0 gr/dL, while 167 (5.3%) had Hb levels >8.0 gr/dL. In addition, 331 (25.3%) of the transfused patients with comorbidities had Hb >8.0 gr/dL. The Hb transfusion thresholds significantly differed across the clinics (p < 0.001), while 21.8% of all medical non-bleeding patients received more than one RBC unit transfusion. Conclusion: A poor adherence with the restrictive transfusion threshold of 7.0 gr/dL was observed. The adoption of a less strict threshold might be a temporary step to allow physicians to become familiar with the program and be informed on the safety and advantages of the restrictive transfusion strategy.

Should We Stop Collecting the Preoperative Autologous Blood before Bone Marrow Harvest?

Preoperative autologous blood donation (PAD) in bone marrow (BM) donors is performed to meet potential post-harvest transfusion needs and to avoid the risk of allogeneic transfusions. We reviewed retrospectively bone marrow harvests in 216 healthy donors during a ten-year period to determine the use of autologous blood. All donors except four had undergone PAD. The initial hemoglobin level of 153 g/L (male donors) and 135 g/L (female donors), respectively, decreased by about 8 g/L after preoperative blood donation and by 23 g/L after bone marrow harvest (medians). Autologous blood was administered to 70% of donors, 30% of the units remained unused. The evaluation of the risk of reaching transfusion threshold (<115 g/L males, <105 g/L females) revealed that donors with initial hemoglobin above 145 g/L and those weighing above 75 kg have minimal risk of requiring blood substitution (about 10%). A larger volume of bone marrow was obtained from male compared to female donors (1300 vs. 1100 mL) because of their higher body weight, which resulted in a higher number of procured nucleated cells (362 vs. 307 × 106/kg TNC, ns). The donor-recipient weight difference predicted the probability of sufficient collection. Only 1.5% of donors weighing ≥ 20 kg more than recipients failed to reach ≥3 × 108/kg TNC recipient. Our findings affirm previous data that PAD is unnecessary for healthy marrow donors and may be indicated individually after considering the pre-collection hemoglobin level, donor and recipient weight, and expected blood loss. Reasonable substitution cut-offs have to be set together with clinical symptom evaluation. The effective use of PAD also requires an adequate time interval between PAD and BM harvest.