The use of predictive modelling to determine the likelihood of donor return during the COVID-19 pandemic.

Artificial intelligence (AI) uses sophisticated algorithms to "learn" from large volumes of data. This could be used to optimise recruitment of blood donors through predictive modelling of future blood supply, based on previous donation and transfusion demand. We sought to assess utilisation of predictive modelling and AI blood establishments (BE) and conducted predictive modelling to illustrate its use. A BE survey of data modelling and AI was disseminated to the International Society of Blood transfusion members. Additional anonymzed data were obtained from Italy, Singapore and the United States (US) to build predictive models for each region, using January 2018 through August 2019 data to determine likelihood of donation within a prescribed number of months. Donations were from March 2020 to June 2021. Ninety ISBT members responded to the survey. Predictive modelling was used by 33 (36.7%) respondents and 12 (13.3%) reported AI use. Forty-four (48.9%) indicated their institutions do not utilise predictive modelling nor AI to predict transfusion demand or optimise donor recruitment. In the predictive modelling case study involving three sites, the most important variable for predicting donor return was number of previous donations for Italy and the US, and donation frequency for Singapore. Donation rates declined in each region during COVID-19. Throughout the observation period the predictive model was able to consistently identify those individuals who were most likely to return to donate blood. The majority of BE do not use predictive modelling and AI. The effectiveness of predictive model in determining likelihood of donor return was validated; implementation of this method could prove useful for BE operations.

A randomized cross-over study of cryopreserved platelets in prophylactic transfusions of thrombocytopenic patients.

BACKGROUND: The short shelf-life of liquid-stored platelets (LP) at 20-24°C poses shortage and wastage challenges. Cryopreserved platelets have significantly extended shelf-life, and were safe and efficacious for therapeutic transfusions of bleeding patients in the Afghanistan conflict and phase 2 randomized studies. Although hematology patients account for half of platelets demand, there is no randomized study on prophylactic cryopreserved platelet transfusions in them. METHODS: We performed a phase 1b/2a randomized cross-over study comparing the safety and efficacy of cryopreserved buffy coat-derived pooled platelets (CP) to LP in the prophylactic transfusions of thrombocytopenic hematology patients. RESULTS: A total of 18 adults were randomly assigned 1:1 to CP and LP for their first thrombocytopenic period (TP) of up to 28-days. A total of 14 crossed over to the other platelet-arm for the second TP. Overall, 17 subjects received 51 CP and 15 received 52 LP. CP-arm had more treatment emergent adverse event (29.4% vs. 13.3% of subjects, 9.8% vs. 3.8% of transfusions) than LP-arm but all were mild. No thromboembolism was observed. Both arms had similar bleeding rates (23.5% vs. 26.7% of subjects) which were all mild. Subjects in CP-arm had lower average corrected count increments than LP-arm (mean [SD] 5.6 [4.20] vs. 22.6 [9.68] ×109 /L at 1-4 h, p < .001; 5.3 [4.84] vs. 18.2 [9.52] ×109 /L at 18-30 h, p < .001). All TEG parameters at 1-4 h and maximum amplitude (MA) at 18-30 h improved from baseline post-CP transfusion (p < .05) though improvements in K-time and MA were lower than LP (p < .05). DISCUSSION: During shortages, CP may supplement LP in prophylactic transfusions of thrombocytopenic patients.

International Society of Blood Transfusion survey of experiences of blood banks and transfusion services during the COVID-19 pandemic.

BACKGROUND AND OBJECTIVES: The coronavirus disease 2019 (COVID-19) pandemic has impacted blood systems worldwide. Challenges included maintaining blood supplies and initiating the collection and use of COVID-19 convalescent plasma (CCP). Sharing information on the challenges can help improve blood collection and utilization. MATERIALS AND METHODS: A survey questionnaire was distributed to International Society of Blood Transfusion members in 95 countries. We recorded respondents' demographic information, impacts on the blood supply, CCP collection and use, transfusion demands and operational challenges. RESULTS: Eighty-two responses from 42 countries, including 24 low- and middle-income countries, were analysed. Participants worked in national (26.8%) and regional (26.8%) blood establishments and hospital-based (42.7%) institutions. CCP collection and transfusion were reported by 63% and 36.6% of respondents, respectively. Decreases in blood donations occurred in 70.6% of collecting facilities. Despite safety measures and recruitment strategies, donor fear and refusal of institutions to host blood drives were major contributing factors. Almost half of respondents working at transfusion medicine services were from large hospitals with over 10,000 red cell transfusions per year, and 76.8% of those hospitals experienced blood shortages. Practices varied in accepting donors for blood or CCP donations after a history of COVID-19 infection, CCP transfusion, or vaccination. Operational challenges included loss of staff, increased workloads and delays in reagent supplies. Almost half of the institutions modified their disaster plans during the pandemic. CONCLUSION: The challenges faced by blood systems during the COVID-19 pandemic highlight the need for guidance, harmonization, and strengthening of the preparedness and the capacity of blood systems against future infectious threats.

Non-transfusion dependent thalassemia is independently associated with higher alloimmunization risk than transfusion dependent thalassemia and would benefit the most from extended red cell antigen-matching.

BACKGROUND: Alloimmunization prevalence is conventionally used to identify RBCs alloimmunization risk factors among thalassemia patients, but it may be confounded by differences in transfusion exposure especially between non-transfusion dependent thalassemia (NTDT) and transfusion dependent thalassemia (TDT) patients. To better identify thalassemia patients with high alloimmunization risks, we used cumulative incidence of first alloimmunization as a function of RBCs transfused to compare alloimmunization risks between TDT and NTDT and to evaluate other risk factors. We also proposed practical strategies to prevent alloimmunization in thalassemia. STUDY DESIGN AND METHODS: Adult TDT and NTDT patients who had received ≥2 transfusions and no alloimmunization before their first transfusion were included. Alloimmunization was defined as the development of clinically significant alloantibodies. We estimated the first alloimmunization incidence from transfusion by Kaplan-Meier analysis with the horizontal axis expressed as cumulative non-antigen-matched RBC units transfused. We compared this incidence between TDT and NTDT, and analyzed for other alloimmunization risk factors and the alloantibody specificities/frequencies. RESULTS: The alloimmunization prevalence was similar between TDT and NTDT (27% vs. 30% respectively, p = .726). However, for the same transfusion exposure, NTDT had higher alloimmunization incidence than TDT (hazard ratio 8.59, 95% confidence interval [2.25-32.74], p = .002), independent of age at first transfusion and last follow-up, gender, and splenectomy. Anti-E, anti-c, anti-Mia , and anti-Jka were most frequent. DISCUSSION: NTDT has the highest alloimmunization risk and would benefit the most from extended RBC antigen-matching, especially C, c, E, and e. Other blood group antigen-matching should be guided by the patient/donor disparities and alloantibody frequencies in different populations.

Clinical and health-related quality of life outcomes of transfusion-dependent thalassaemia patients in Singapore.

INTRODUCTION: Transfusion-dependent thalassaemia is associated with complications related to iron overload from frequent red cell transfusions which affect quality of life. We collected data on the clinical outcomes, complications, socioeconomic status and health-related quality of life (HRQoL) of transfusion-dependent thalassaemia patients in Singapore, and analysed the associations between clinical and socioeconomic factors with development of transfusion-related complications and HRQoL scores. MATERIALS AND METHODS: This was a cross-sectional study of transfusion-dependent thalassaemia patients treated at four major public hospitals in Singapore. Clinical information was obtained from retrospective reviews of medical records. Socioeconomic data and patient-reported compliance to iron chelators were obtained from prospective interviews of patients or caregivers using a questionnaire. A validated, disease-specific HRQoL instrument, the TranQOL, was administered to patients and caregivers during a routine clinic or transfusion visit. RESULTS: Liver iron loading was the most common transfusion-related complication and occurred in 79% of patients. Cardiac iron loading was noted in 28.3% and endocrine complications were present in 34.2%. Liver iron loading was significantly associated with higher mean ferritin level. Cardiac iron loading was significantly associated with increasing age, higher mean ferritin level and type of iron chelator. Endocrine complications were associated with increasing age, higher mean ferritin level, type of iron chelator and poorer patient-reported compliance to iron chelators. The lowest TranQOL scores were reported by caregiver parents of patients aged less than 18 years. Lower TranQOL scores were significantly associated with increasing age, especially in the 31-50 age cohort, and with reception of social assistance. CONCLUSION: The main morbidities noted in transfusion-dependent thalassaemia patients in Singapore are from complications associated with iron loading. The cohort of older thalassaemia patients aged 31-50 experienced significantly higher rates of cardiac iron loading, endocrine complications and lower TranQOL scores compared to younger age cohorts.

Guidance for the procurement of COVID-19 convalescent plasma: differences between high- and low-middle-income countries.

BACKGROUND AND OBJECTIVES: COVID-19 convalescent plasma (CCP) has been used, predominantly in high-income countries (HICs) to treat COVID-19; available data suggest the safety and efficacy of use. We sought to develop guidance for procurement and use of CCP, particularly in low- and middle-income countries (LMICs) for which data are lacking. MATERIALS AND METHODS: A multidisciplinary, geographically representative group of individuals with expertise spanning transfusion medicine, infectious diseases and haematology was tasked with the development of a guidance document for CCP, drawing on expert opinion, survey of group members and review of available evidence. Three subgroups (i.e. donor, product and patient) were established based on self-identified expertise and interest. Here, the donor and product-related challenges are summarized and contrasted between HICs and LMICs with a view to guide related practices. RESULTS: The challenges to advance CCP therapy are different between HICs and LMICs. Early challenges in HICs related to recruitment and qualification of sufficient donors to meet the growing demand. Antibody testing also posed a specific obstacle given lack of standardization, variable performance of the assays in use and uncertain interpretation of results. In LMICs, an extant transfusion deficit, suboptimal models of donor recruitment (e.g. reliance on replacement and paid donors), limited laboratory capacity for pre-donation qualification and operational considerations could impede wide adoption. CONCLUSION: There has been wide-scale adoption of CCP in many HICs, which could increase if clinical trials show efficacy of use. By contrast, LMICs, having received little attention, require locally applicable strategies for adoption of CCP.

Getting patient blood management Pillar 1 right in the Asia-Pacific: a call for action.

Preoperative anaemia is common in the Asia-Pacific. Iron deficiency anaemia (IDA) is a risk factor that can be addressed under patient blood management (PBM) Pillar 1, leading to reduced morbidity and mortality. We examined PBM implementation under four different healthcare systems, identified challenges and proposed several measures: (a) Test for anaemia once patients are scheduled for surgery. (b) Inform patients about risks of preoperative anaemia and benefits of treatment. (c) Treat IDA and replenish iron stores before surgery, using intravenous iron when oral treatment is ineffective, not tolerated or when rapid iron replenishment is needed; transfusion should not be the default management. (d) Harness support from multiple medical disciplines and relevant bodies to promote PBM implementation. (e) Demonstrate better outcomes and cost savings from reduced mortality and morbidity. Although PBM implementation may seem complex and daunting, it is feasible to start small. Implementing PBM Pillar 1, particularly in preoperative patients, is a sensible first step regardless of the healthcare setting.

Ex vivo haemostatic capacity of plasma upon thawing and beyond: a comparison between fresh frozen plasma (FFP) and frozen plasma prepared from whole blood stored at room temperature up to 24 hours postcollection (RTFP24).

BACKGROUND/OBJECTIVES: We compared the ex vivo haemostatic capacity of RTFP24 with FFP upon thawing and >24 h post-thaw. We included thrombin generation (TG) as few studies had compared global haemostatic function, and most did not directly compare RTFP24 with FFP >24 h post-thaw. MATERIALS/METHODS: Twenty units each of RTFP24 and FFP were measured for coagulation factors and thrombin generation upon thawing (D0) and 4 days post-thaw (D4). Labile factors were also measured from D1 to D3. 10 single cryoprecipitate units were each prepared from FFP and RTFP24, and measured for FXIII, FVIII and fibrinogen at D0. RESULTS: At D0, RTFP24 was comparable to FFP except for lower FV, protein S, endogenous thrombin potential (ETP) and higher FXIII. These differences were likely clinically insignificant since 95% and 80% of RTFP24 met our laboratory's reference ranges for FV/protein S and ETP, respectively. There were no differences between RTFP24- and FFP-derived cryoprecipitate. At D4, RTFP24 was comparable to FFP except for lower FV, ETP, and higher FXI and FXIII. More RTFP24 than FFP had ETP lower than our laboratory's reference range (45% vs 15%). Multiple coagulation factors and all TG parameters declined from their respective baselines. The percentage declines were comparable or less in RTFP24, except for protein C, fibrinogen and time to peak. CONCLUSION: RTFP24 and FFP, and their derived cryoprecipitate have comparable haemostatic capacity upon thawing. RTFP24 has poorer TG potential than FFP >24 h post-thaw, not supporting universal extension of RTFP24's shelf life except to facilitate urgent transfusions for massive haemorrhage.

Role of platelets in thrombin generation amongst patients with non-transfusion-dependent thalassaemia.

Non-transfusion-dependent thalassaemia (NTDT) is associated with a hypercoagulable state with thrombotic risk highest after splenectomy. Various mechanisms have been proposed. Although an antiplatelet agent is commonly recommended as thromboprophylaxis in NTDT, the role of platelets contributing to this hypercoagulable state is not well-defined. This study aims to evaluate the role of platelets contributing to hypercoagulability in NTDT patients using thrombin generation (TG). Platelet-rich (PRP) and platelet-poor plasma (PPP) were collected from NTDT patients (n = 30) and normal controls (n = 20) for TG measurement and compared. Controls had higher endogenous thrombin potential (ETP) in PPP (1204.97 nM.min vs 911.62 nM.min, p < 0.001) and PRP (1424.23 nM.min vs 983.99 nM.min, p < 0.001) than patients. Patients' mean normalized ETP ratio [{PRP ETP (patient)/PPP ETP (patient)}/{mean PPP ETP (controls)/mean PPP ETP (controls)}], demonstrated that the presence of platelet does not alter ETP (mean ratio 0.97, 95% CI 0.93-1.02, equivalence defined as 10%). Types of thalassaemia, splenectomy, and severity of liver iron overload did not significantly influence patients' ETP in PPP and PRP by multivariate analysis. Platelets did not increase the TG potential of NTDT patients. Instead of being hypercoagulable, our NTDT patients were hypocoagulable by ETP measurement, although this could not be conclusively demonstrated to correlate with their iron overloading state giving rise to reduced synthesis of coagulation factors. The guideline recommendations for thromboprophylaxis with antiplatelet agents in similar NTDT patients should be re-examined.

Prevalence of preoperative anemia, abnormal mean corpuscular volume and red cell distribution width among surgical patients in Singapore, and their influence on one year mortality.

INTRODUCTION: Preoperative anemia and high red cell distribution width (RDW) are associated with higher perioperative mortality. Conditions with high RDW levels can be categorized by mean corpuscular volume (MCV). The relationship between RDW, anemia and MCV may explain causality between high RDW levels and outcomes. We aim to establish the prevalence of preoperative anemia and distribution of RDW and MCV among pre-surgical patients in Singapore. In addition, we aim to investigate the association between preoperative anemia, RDW and MCV levels with one-year mortality after surgery. METHODS: Retrospective review of 97,443 patients aged > = 18 years who underwent cardiac and non-cardiac surgeries under anesthesia between January 2012 and October 2016. Patient demographics, comorbidities, priority of surgery, surgical risk classification, perioperative transfusion, preoperative hemoglobin, RDW, MCV were collected. WHO anemia classification was used. High RDW was defined as >15.7%. Multivariate regression analyses were done to identify independent risk factors for mild or moderate/severe anemia and high RDW (>15.7). Multivariate cox regression analysis was done to determine the effect of preoperative anemia, abnormal RDW and MCV values on 1-year mortality. RESULTS: Our cohort comprised of 94.7% non-cardiac and 5.3% cardiac surgeries. 88.7% of patients achieved 1 year follow-up. Anemia prevalence was 27.8%-mild anemia 15.3%, moderate anemia 12.0% and severe anemia 0.5%. One-year mortality was 3.5%. Anemia increased with age in males, while in females, anemia was more prevalent between 18-49 years and > = 70 years. Most anemics were normocytic. Normocytosis and macrocytosis increased with age, while microcytosis decreased with age. Older age, male gender, higher ASA-PS score, anemia (mild- aHR 1.98; moderate/severe aHR 2.86), macrocytosis (aHR 1.47), high RDW (aHR 2.34), moderate-high risk surgery and emergency surgery were associated with higher hazard ratios of one-year mortality. DISCUSSION: Preoperative anemia is common. Anemia, macrocytosis and high RDW increases one year mortality.