How to begin using social media for transfusion medicine education.

This article highlights fundamentals that are important for the transfusion medicine educator to understand about social media. Several examples of personal practical application are shared. Finally, the potential future state of social media will be discussed. In the spirit of a growth mindset, please suspend any previous judgements about social media and allow yourself to consider the possibility of using social media with your transfusion education.

Prozone rates in the solid-phase platelet crossmatch assay and correlation with class I HLA antibody levels.

BACKGROUND: Solid-phase platelet crossmatch (PXM) testing is used to help manage patients with platelet transfusion-refractoriness. Recently, we published the first report of false-negative PXM results from prozone effect that was mitigated using sample dilution. This study aimed to describe the prevalence of PXM prozone effect and the levels of class I HLA antibodies (HLA-Abs) associated with positive PXM results and with false-negative PXM results from prozone effect. STUDY DESIGN AND METHODS: A cross-sectional study of patients undergoing PXM testing from July 2019 through December 2020 was performed. All PXM tests were run simultaneously using undiluted and 1:4 diluted patient plasma. Prozone effect was defined as a negative PXM result using undiluted patient plasma but a positive PXM result using 1:4 diluted patient plasma. RESULTS: Among 59 patients, 830 individual ABO-compatible PXM results yielded an overall positivity rate of 25.8% (214/830) and a false-negative rate from prozone effect of 4.7% (10/214). Among the 28 patients with class I HLA-Ab testing and no other anti-platelet antibodies, maximum HLA-Ab mean fluorescence intensity (MFI) was significantly associated with a positive PXM result (p < .0001; AUC approx. 0.9) and categorized into negative (<3700), indeterminate (3700-10300), and positive (>10300) maximum HLA-Ab MFI zones. Maximum HLA-Ab MFI, however, was not associated with prozone effect (p = .17; AUC approx. 0.6). DISCUSSION: While there is a strong predictive association between class I HLA-Ab levels and positive PXM results, PXM prozone effect is a common occurrence not associated with class I HLA-Ab levels, so additional testing with diluted samples should be considered.

Daratumumab interference in flow cytometric anti-granulocyte antibody testing can be overcome using non-human blocking antibodies.

BACKGROUND: Daratumumab (DARA), a human IgG1K monoclonal antibody targeting CD38, is used to treat refractory multiple myeloma patients. CD38 is expressed on many cell types (RBCs, granulocytes, lymphocytes, etc.), and thus, DARA can interfere with serological tests. Information regarding how DARA affects anti-granulocyte antibody (AGA) testing and optimal neutralization of DARA will help laboratories perform accurate testing. METHODS: Screening of AGA was performed by the granulocyte agglutination test (GAT) and the flow cytometric granulocyte immunofluorescence test (Flow-GIFT). Samples were tested from patients on DARA (n = 7), non-transfused blood donors (healthy controls, n = 7) and AGA reactive samples (positive controls, n = 5). Two neutralization experiments, CD38 removal with DTT and DARA epitope blockage with mouse anti-CD38, were evaluated. RESULTS: Positive reactivity of human IgG binding was observed in 5/7 DARA cases when tested by Flow-GIFT; however, all 7 cases had negative GAT agglutination results. Further studies by Flow-GIFT revealed DARA concentrations >0·63 µg/ml bound to granulocytes. DARA binding was negated by DTT though a reduced Flow-GIFT sensitivity was observed in positive control samples due to increased background detection of human IgG. Mouse anti-CD38 neutralized the detection of human IgG observed in DARA-treated patient serum without effecting controls. CONCLUSION: We established that DARA can interfere with AGA testing, leading to false positive Flow-GIFT results without causing GAT agglutination. DTT treatment increased background binding of secondary antibodies causing a decrease in Flow-GIFT sensitivity. In comparison, blockage of the DARA binding epitope using mouse anti-CD38 antibody was effective in neutralizing DARA interference while maintaining Flow-GIFT sensitivity.

COVID-19 Mortality Prediction From Deep Learning in a Large Multistate Electronic Health Record and Laboratory Information System Data Set: Algorithm Development and Validation.

BACKGROUND: COVID-19 is caused by the SARS-CoV-2 virus and has strikingly heterogeneous clinical manifestations, with most individuals contracting mild disease but a substantial minority experiencing fulminant cardiopulmonary symptoms or death. The clinical covariates and the laboratory tests performed on a patient provide robust statistics to guide clinical treatment. Deep learning approaches on a data set of this nature enable patient stratification and provide methods to guide clinical treatment. OBJECTIVE: Here, we report on the development and prospective validation of a state-of-the-art machine learning model to provide mortality prediction shortly after confirmation of SARS-CoV-2 infection in the Mayo Clinic patient population. METHODS: We retrospectively constructed one of the largest reported and most geographically diverse laboratory information system and electronic health record of COVID-19 data sets in the published literature, which included 11,807 patients residing in 41 states of the United States of America and treated at medical sites across 5 states in 3 time zones. Traditional machine learning models were evaluated independently as well as in a stacked learner approach by using AutoGluon, and various recurrent neural network architectures were considered. The traditional machine learning models were implemented using the AutoGluon-Tabular framework, whereas the recurrent neural networks utilized the TensorFlow Keras framework. We trained these models to operate solely using routine laboratory measurements and clinical covariates available within 72 hours of a patient's first positive COVID-19 nucleic acid test result. RESULTS: The GRU-D recurrent neural network achieved peak cross-validation performance with 0.938 (SE 0.004) as the area under the receiver operating characteristic (AUROC) curve. This model retained strong performance by reducing the follow-up time to 12 hours (0.916 [SE 0.005] AUROC), and the leave-one-out feature importance analysis indicated that the most independently valuable features were age, Charlson comorbidity index, minimum oxygen saturation, fibrinogen level, and serum iron level. In the prospective testing cohort, this model provided an AUROC of 0.901 and a statistically significant difference in survival (P<.001, hazard ratio for those predicted to survive, 95% CI 0.043-0.106). CONCLUSIONS: Our deep learning approach using GRU-D provides an alert system to flag mortality for COVID-19-positive patients by using clinical covariates and laboratory values within a 72-hour window after the first positive nucleic acid test result.

Development and performance characteristics of Platelet Virtual Crossmatch (PLT VXM), a software application for the evaluation and management of platelet transfusion-refractory patients.

BACKGROUND: Platelet (PLT) transfusion refractoriness increases bleeding complications, hospital stays, and PLT inventory usage. Immune-mediated refractoriness can be evaluated for using a physical PLT crossmatch with ABO-compatible inventory and, if positive, managed with HLA-compatible PLT inventory and donors. Manual completion of these complex tasks can be time-consuming and potentially error-prone. This study was conducted to determine if a Web-based software application could improve process efficiency and accuracy. STUDY DESIGN AND METHODS: Workflow analysis was performed to identify process, data, and analytic requirements for a software application for three PLT transfusion-refractoriness associated tasks: (a) physical PLT crossmatch inventory selection, (b) HLA-compatible inventory selection, and (c) HLA-compatible donor selection. After software application development, a comparison study was performed over 10 consecutive days, with each task performed manually and with the software application (Platelet Virtual Crossmatch [PLT VXM]) for a different unique immune-mediated PLT transfusion-refractory recipient. Task completion time, number of incompatible units/donors presented, and number of documentation errors were compared. RESULTS: PLT VXM is a Web-based software application developed using R and the Shiny Web application framework. PLT VXM significantly reduced median task completion times by 4.5 (49%), 11.2 (79%), and 59.1 minutes (94%), respectively. PLT VXM did not present any incompatible PLT units or donors for user consideration. PLT VXM also had a lower number of documentation errors than the manual process, and none of these documentation errors were software generated. CONCLUSION: Computer-aided evaluation and management of immune-mediated PLT transfusion-refractory recipients can significantly improve workflow and reduce manual errors in this complex process.

Implementation of a patient blood management program in hematopoietic stem cell transplantation (Editorial, p. 2763).

BACKGROUND: Recipients of hematopoietic stem cell transplantation (HSCT) are among the highest consumers of allogeneic red blood cell (RBC) and platelet (PLT) components. The impact of patient blood management (PBM) efforts on HSCT recipients is poorly understood. STUDY DESIGN AND METHODS: This observational study assessed changes in blood product use and patient-centered outcomes before and after implementing a multidisciplinary PBM program for patients undergoing HSCT at a large academic medical center. The pre-PBM cohort was treated from January 1 through September 31, 2013; the post-PBM cohort was treated from January 1 through September 31, 2015. RESULTS: We identified 708 patients; 284 of 352 (80.7%) in the pre-PBM group and 225 of 356 (63.2%) in the post-PBM group received allogeneic RBCs (p < 0.001). Median (interquartile range [IQR]) RBC volumes were higher before PBM than after PBM (3 [2-4] units vs. 2 [1-4] units; p = 0.004). A total of 259 of 284 pre-PBM patients (91.2%) and 57 of 225 (25.3%) post-PBM patients received RBC transfusions when hemoglobin levels were more than 7 g/dL (p < 0.001). The median (IQR) PLT transfusion quantities was 3 (2-5) units for pre-PBM patients and 2 (1-4) units for post-PBM patients (p < 0.001). For patients with PLT counts of more than 10 × 109 /L, a total of 1219 PLT units (73.4%) were transfused before PBM and 691 units (48.8%) were transfused after PBM (p < 0.001). Estimated transfusion expenditures were reduced by $617,152 (18.3%). We noted no differences in clinical outcomes or transfusion-related adverse events. CONCLUSION: Patient blood management implementation for HSCT recipients was associated with marked reductions in allogeneic RBC and PLT transfusions and decreased transfusion-related costs with no detrimental impact on clinical outcomes.

Extracorporeal Photopheresis Improves Survival in Hematopoietic Cell Transplant Patients with Bronchiolitis Obliterans Syndrome without Significantly Impacting Measured Pulmonary Functions.

We carried out the first matched retrospective cohort study aimed at studying the safety and efficacy of extracorporeal photopheresis (ECP) for bronchiolitis obliterans syndrome (BOS) after allogeneic hematopoietic cell transplantation (HCT). Medical records of 1325 consecutive adult patients who underwent HCT between 2005 and 2015 were reviewed. Seventy-four patients (median age, 51 years) with a diagnosis of BOS were included in the study. After propensity-score matching for BOS severity, 26 patients who underwent ≥3 months of ECP were matched to 26 non-ECP-treated patients, who were assigned an index date corresponding to the ECP start date for their matched pairs. The rate of decline in FEV1 percentage predicted (FEV1PP) decreased after ECP initiation (and after index date in the non-ECP group), with no significant difference between the 2 groups (P = .33). On a multivariable analysis that included baseline transplant and pulmonary function test variables, matched related donor HCT (HR, .1; 95% CI, .03 to .5; P = .002), ECP (HR, .1; 95% CI, .01 to .3; P = .001), and slower rate of decline in FEV1PP before the ECP/index date (HR, .7; 95% CI, .6 to .8; P = .001) were associated with a better overall survival. At last follow-up, non-ECP-treated patients were more likely to be on >5 mg daily dose of prednisone (54% versus 23%; P = .04) and had a greater decline in their Karnofsky performance score (mean difference, -9.5 versus -1.6; P = .06) compared with ECP-treated-patients. In conclusion, compared with other BOS-directed therapies, ECP was found to improve survival in HCT patients with BOS, without significantly impacting measured pulmonary functions. These findings need prospective validation in a larger patient cohort.

Changes in International Normalized Ratios After Plasma Transfusion of Varying Doses in Unique Clinical Environments.

BACKGROUND: Plasma transfusion is commonly performed for the correction of abnormal coagulation screening tests. The goal of this investigation was to assess the relationship between the dose of plasma administered and changes in coagulation test results in a large and diverse cohort of patients with varying levels of coagulation abnormalities and comorbid disease and in a variety of clinical settings. METHODS: In this single-center historical cohort study, all plasma transfusion episodes in adult patients with abnormal coagulation screening tests were extracted between 2011 and 2015. The primary outcome was the proportion of patients attaining normal posttransfusion international normalized ratio (INR ≤ 1.1) with secondary outcomes including the proportion of patients attaining partial normalization of INR (INR ≤ 1.5) or at least 50% normalization in pretransfusion values with respect to an INR of 1.1. RESULTS: In total, 6779 unique patients received plasma with a median (quartiles) pretransfusion INR of 1.9 (1.6-2.5) and a median transfusion volume of 2 (2-3) units. The majority (85%) of transfusions occurred perioperatively, with 20% of transfusions administered prophylactically before a procedure. The median decrease in INR was 0.4 (0.2-0.8). Complete INR normalization was obtained in 12%. Reductions in INR were modest with pretransfusion INR values <3. Patients receiving ≥3 units of plasma were more likely to achieve at least 50% normalization in INR than those receiving ≤2 units (68% vs 60%; P < .001). CONCLUSIONS: Changes in INR after plasma transfusion were modest at typically used clinical doses, particularly in those with less severely deranged baseline coagulation screening tests. Further studies are necessary to assess the relationships between plasma-mediated changes in INR and clinical outcomes.

How do I manage the platelet transfusion-refractory patient?

BACKGROUND: Platelet transfusion-refractoriness is a challenging and expensive clinical scenario seen most often in patients with hematologic malignancies. Although the majority of platelet transfusion-refractory cases are due to nonimmune causes, a significant minority are caused by alloimmunization against Class I human leukocyte antigens (HLAs) or human platelet antigens (HPAs). Such platelet transfusion-refractory patients can be effectively managed with appropriate antigen-negative products. STUDY DESIGN AND METHODS: Our institution has developed a diagnostic and management algorithm for the platelet transfusion-refractory patient with an early focus on identifying those cases caused by immune-mediated factors. Using physical platelet cross-matches to initially classify platelet transfusion-refractory patients as immune-mediated or not, cross-match-compatible inventory is then provided to immune-mediated patients, whereas subsequent HLA (with or without HPA) testing is performed. RESULTS: Our blood donor program performs Class I HLA typing of all repeat platelet donors to facilitate the identification of antigen-negative platelet units (virtual cross-matching) as well as the recruitment of HLA-matched donors. The platelet transfusion-refractoriness algorithm realizes an initial net cost savings once two apheresis platelets are saved from use for each newly identified, immune-mediated platelet transfusion-refractory patient. CONCLUSION: An algorithm utilizing physical platelet cross-matches, Class I HLA and HPA antibody testing, and upfront Class I HLA typing of platelet donors leads to overall resource savings and improved clinical management for platelet transfusion-refractory patients.

Clinical outcomes of HLA-DPB1 mismatches in 10/10 HLA-matched unrelated donor-recipient pairs undergoing allogeneic stem cell transplant.

OBJECTIVE: HLA-DPB1 matching may impact allogeneic hematopoietic stem cell transplantation (ASCT) outcomes; however, this locus is not in linkage disequilibrium with the remainder of the HLA genes. After classifying HLA-DPB1 mismatches based on T-cell epitope, avoiding non-permissive mismatches may impact survival. We tested this hypothesis at a single academic institution. METHODS: Retrospective HLA-DPB1 genotyping was performed on 153 adult patients who underwent ASCT and unrelated donors matched for HLA-A, B, C, DRB1, and DQB1 loci (10/10). Using the ImMunoGeneTics/HLA T-cell epitope matching algorithm, mismatch status was classified as permissive or non-permissive. RESULTS: Of 153 donor-recipient pairs, 22 (14.4%) were HLA-DPB1 matches, 64 (42.8%) permissive mismatches, and 67 (43.8%) non-permissive mismatches. DPB1 mismatch increased risk of chronic graft-versus-host disease (cGVHD; RR 2.89 [1.19-9.53], P=.016) compared with DPB1-matched transplants, but there were no differences in overall mortality, risk of relapse, or acute GVHD (aGVHD). Combining matches and permissive mismatches and comparing to non-permissive mismatches, there was no significant difference in overall survival or relapse; however, patients receiving non-permissive mismatched transplants experienced greater risk of aGVHD overall and severe aGVHD (RR 1.66 [1.13-2.44], P=.010 and RR 1.97 [1.10-3.59], P=.024, respectively). CONCLUSION: In this single-center study, HLA-DPB1 matching influenced outcomes of patients undergoing ASCT for hematologic malignancy.

Autoimmune Hemolytic Anemia in Children: Mayo Clinic Experience.

We studied 35 pediatric patients with autoimmune hemolytic anemia seen at Mayo Clinic from 1994 to 2014. The median age was 10.0 years and 65.7% were males. Most had warm antibodies (80.0%) and some secondary to viral (14.3%) or autoimmune disorders (31.4%). Seven (20.0%) patients presented with Evans syndrome, 3 of whom also had common variable immunodeficiency. The median hemoglobin at diagnosis was 6.1 g/dL and 62.8% patients required red cell transfusions. The severity of anemia was worse among children below 10 years (median 5.5 vs. 7.0 g/dL, P=0.01). Steroid was the initial treatment for 88.5% patients, with overall response rate of 82.7% (68.5% complete, 14.2% partial) and median response duration of 10.7 months (range, 0.2 to 129.7+ mo). After median follow-up of 26.6 months, 8 (22.8%) patients relapsed. Salvage treatments included splenectomy, intravenous immunoglobulin, rituximab, and mycophenolate mofetil. Infectious complications occurred in 9 (25.7%) patients and 1 patient died of cytomegalovirus infection. Four patients had cold agglutinin disease and 3 (75.0%) responded to steroids. Autoimmune hemolytic anemia is a rare disorder in pediatric population and most respond well to steroids regardless of the type of antibody. Infectious complications are common and screening for immunodeficiency is recommended among those with Evans syndrome.

Outcomes of plasma exchange in patients with transplant-associated thrombotic microangiopathy based on time of presentation since transplant.

Transplant-associated thrombotic microangiopathy (TA-TMA) is a rare clinical syndrome associated with significant mortality. Although the use of plasma exchange (PE) in TA-TMA continues to be explored, evidence for its efficacy is debated. We performed a single institution, retrospective study to evaluate the efficacy of PE in treating TA-TMA patients. Special attention was given to efficacy in relation to the timing of presentation with TA-TMA since transplant. Thirty-three patients diagnosed with TA-TMA and treated with PE between January 1999 and December 2010 were included in the study. Clinical improvement was seen in eight patients (24%); four patients achieved complete resolution while the remaining four achieved partial resolution. All-cause day-30 and day-100 mortality was 33 and 55%, respectively. There was a trend toward a better outcome (complete/partial) for those presenting ≥ 100 days after transplantation (42%) vs. < 100 days after transplantation (14%; P-value = 0.15). Similarly, those presenting at ≥ 100 days had better, but not significantly, 30-day and 100-day all-cause mortality rates (17 and 33%, respectively) than those presenting at < 100 days (43 and 67%, respectively) (P-value = 0.25 and 0.08, for 30- and 100-day all-cause mortality, respectively). This is the first study looking at the efficacy of PE while considering the time of presentation since transplantation and is one of the largest single institution series of TA-TMA. The overall efficacy of PE is poor; however, patients who present with TA-TMA ≥100 days after transplant may have better outcome and lower mortality.

Routine bacterial screening of apheresis platelets on Day 4 using a rapid test: a 4-year single-center experience.

BACKGROUND: The platelet (PLT) Pan Genera Detection test (PGD) is a rapid bacterial detection system used to screen PLTs for bacterial contamination. We report a single center 46-month experience with secondary screening of apheresis PLTs by PGD testing. STUDY DESIGN AND METHODS: Existing testing records of apheresis PLTs screened by PGD from July 2008 to April 2012 were reviewed. All PLT units were initially screened by routine postcollection culture methods. Secondary screening using PGD was performed for indated PLTs on PLT storage Day 4 and for outdated PLTs on Day 8. RESULTS: A total of 8535 apheresis PLTs were available in inventory during the study period. Of these, 5030 (58.9%) were dispensed and transfused before PGD testing and 3505 (41.1%) underwent PGD testing on Day 4. Twenty-five units tested on Day 4 were PGD initial reactive (0.71%). All were confirmed to be false positive by repeat PGD testing in triplicate (n=20) or by confirmatory culture (n=5). An additional 364 units that were PGD nonreactive on Day 4 were approved for transfusion on Day 6 or Day 7 due to urgent clinical need. A total of 371 outdated units underwent repeat PGD testing before discard on Day 8; all were nonreactive. CONCLUSION: Secondary PGD testing of culture-screened apheresis PLTs results in low yield in a medium-sized transfusion service. Use of PGD testing on Day 4 may allow for extension of the apheresis PLT shelf life to Day 7 for hospitals that face supply constraints.

Clinical impact of KIR haplotypes in 10/10 HLA-matched unrelated donor-recipient pairs undergoing allogeneic hematopoietic stem cell transplantation.

To evaluate the impact of killer immunoglobulin-like receptor (KIR) genotyping in allogeneic hematopoietic stem cell transplantation for myeloid disorders at our institution, retrospective KIR genotyping was performed on 77 patients and their 10/10 matched unrelated donors. In a multivariate model including donor age, HLA-DPB1 permissiveness, and presence of donor KIR B/x, an association with overall survival was observed (p = .047). Within the model, increasing donor age increased risk (RR 1.03 [1.00-1.06]/year, p = .046), while donor KIR and HLA-DPB1 permissiveness were not associated with risk (RR 0.51 [0.26-1.03] and RR 0.68 [0.34-1.36]). Grouping recipients by conditioning regimen or limiting the analysis to recipients of peripheral blood stem cells, no association between donor KIR and survival or relapse was identified. No significant associations were observed between overall survival, relapse, grade III-IV acute, or chronic graft versus host disease and presence of KIR B (B/x), quantity of donor KIR B haplotype motifs, or centromeric KIR type (all p > .05).

Drug-associated thrombotic microangiopathies.

Thrombotic microangiopathy is a pathological process that results in thrombocytopenia, schistocytes, and organ ischemia. A wide variety of causes of thrombotic microangiopathy have been described, each associated with a different pathological stimulus and predisposition toward certain vascular territories. This review will summarize our understanding of key drugs that have been classically and newly associated with thrombotic microangiopathy as well as mention of treatment options focusing on the role of plasma exchange.