Controls, comparator arms, and designs for critical care comparative effectiveness research: It's complicated.

BACKGROUND: Comparative effectiveness research is meant to determine which commonly employed medical interventions are most beneficial, least harmful, and/or most costly in a real-world setting. While the objectives for comparative effectiveness research are clear, the field has failed to develop either a uniform definition of comparative effectiveness research or an appropriate set of recommendations to provide standards for the design of critical care comparative effectiveness research trials, spurring controversy in recent years. The insertion of non-representative control and/or comparator arm subjects into critical care comparative effectiveness research trials can threaten trial subjects' safety. Nonetheless, the broader scientific community does not always appreciate the importance of defining and maintaining critical care practices during a trial, especially when vulnerable, critically ill populations are studied. Consequently, critical care comparative effectiveness research trials sometimes lack properly constructed control or active comparator arms altogether and/or suffer from the inclusion of "unusual critical care" that may adversely affect groups enrolled in one or more arms. This oversight has led to critical care comparative effectiveness research trial designs that impair informed consent, confound interpretation of trial results, and increase the risk of harm for trial participants. METHODS/EXAMPLES: We propose a novel approach to performing critical care comparative effectiveness research trials that mandates the documentation of critical care practices prior to trial initiation. We also classify the most common types of critical care comparative effectiveness research trials, as well as the most frequent errors in trial design. We present examples of these design flaws drawn from past and recently published trials as well as examples of trials that avoided those errors. Finally, we summarize strategies employed successfully in well-designed trials, in hopes of suggesting a comprehensive standard for the field. CONCLUSION: Flawed critical care comparative effectiveness research trial designs can lead to unsound trial conclusions, compromise informed consent, and increase risks to research subjects, undermining the major goal of comparative effectiveness research: to inform current practice. Well-constructed control and comparator arms comprise indispensable elements of critical care comparative effectiveness research trials, key to improving the trials' safety and to generating trial results likely to improve patient outcomes in clinical practice.

Modeling current practices in critical care comparative effectiveness research.

Objective: To determine whether contemporaneous practices are adequately represented in recent critical care comparative effectiveness research studies. Design: All critical care comparative effectiveness research trials published in the New England Journal of Medicine from April 2019 to March 2020 were identified. To examine studies published in other high impact medical journals during the same period, such trials were subsequently also identified in the Journal of the American Medical Association and The Lancet. All cited sources were reviewed, and the medical literature was searched to find studies describing contemporary practices. Then, the designated control group or the comparable therapies studied were examined to determine if they represented contemporaneous critical care practices as described in the medical literature. Results: Twenty-five of 332 randomised clinical trials published in these three journals during this 1-year period described critical care comparative effectiveness research that met our inclusion criteria. Seventeen characterised current practices before enrolment (using surveys, observational studies and guidelines) and then incorporated current practices into one or more study arm. In the other eight, usual care arms appeared insufficient. Four of these trials randomly assigned patients to one of two fixed approaches at either end of a range of usually titrated care. However, due to randomisation, different subgroups within each arm received care that was inappropriate for their specific clinical conditions. In the other four of these trials, common practices influencing treatment choice were not reflected in the trial design, despite a prior effort to characterise usual care. Conclusion: One-third of critical care comparative effectiveness research trials published in widely read medical journals during a recent year did not include a designated control arm or comparable therapies representative of contemporary practices. Failure to incorporate contemporary practices into critical care comparative effectiveness trials appears to be a widespread design weakness.

Mechanistic insights into cell-free hemoglobin-induced injury during septic shock.

Cell-free hemoglobin (CFH) levels are elevated in septic shock and are higher in nonsurvivors. Whether CFH is only a marker of sepsis severity or is involved in pathogenesis is unknown. This study aimed to investigate whether CFH worsens sepsis-associated injuries and to determine potential mechanisms of harm. Fifty-one, 10-12 kg purpose-bred beagles were randomized to receive Staphylococcus aureus intrapulmonary challenges or saline followed by CFH infusions (oxyhemoglobin >80%) or placebo. Animals received antibiotics and intensive care support for 96 h. CFH significantly increased mean pulmonary arterial pressures and right ventricular afterload in both septic and nonseptic animals, effects that were significantly greater in nonsurvivors. These findings are consistent with CFH-associated nitric oxide (NO) scavenging and were associated with significantly depressed cardiac function, and worsened shock, lactate levels, metabolic acidosis, and multiorgan failure. In septic animals only, CFH administration significantly increased mean alveolar-arterial oxygenation gradients, also to a significantly greater degree in nonsurvivors. CFH-associated iron levels were significantly suppressed in infected animals, suggesting that bacterial iron uptake worsened pneumonia. Notably, cytokine levels were similar in survivors and nonsurvivors and were not predictive of outcome. In the absence and presence of infection, CFH infusions resulted in pulmonary hypertension, cardiogenic shock, and multiorgan failure, likely through NO scavenging. In the presence of infection alone, CFH infusions worsened oxygen exchange and lung injury, presumably by supplying iron that promoted bacterial growth. CFH elevation, a known consequence of clinical septic shock, adversely impacts sepsis outcomes through more than one mechanism, and is a biologically plausible, nonantibiotic, noncytokine target for therapeutic intervention.NEW & NOTEWORTHY Cell-free hemoglobin (CFH) elevations are a known consequence of clinical sepsis. Using a two-by-two factorial design and extensive physiological and biochemical evidence, we found a direct mechanism of injury related to nitric oxide scavenging leading to pulmonary hypertension increasing right heart afterload, depressed cardiac function, worsening circulatory failure, and death, as well as an indirect mechanism related to iron toxicity. These discoveries alter conventional thinking about septic shock pathogenesis and provide novel therapeutic approaches.

An open-source python library for detection of known and novel Kell, Duffy and Kidd variants from exome sequencing.

BACKGROUND AND OBJECTIVES: Next generation sequencing (NGS) has promising applications in transfusion medicine. Exome sequencing (ES) is increasingly used in the clinical setting, and blood group interpretation is an additional value that could be extracted from existing data sets. We provide the first release of an open-source software tailored for this purpose and describe its validation with three blood group systems. MATERIALS AND METHODS: The DTM-Tools algorithm was designed and used to analyse 1018 ES NGS files from the ClinSeq® cohort. Predictions were correlated with serology for 5 antigens in a subset of 108 blood samples. Discrepancies were investigated with alternative phenotyping and genotyping methods, including a long-read NGS platform. RESULTS: Of 116 genomic variants queried, those corresponding to 18 known KEL, FY and JK alleles were identified in this cohort. 596 additional exonic variants were identified KEL, ACKR1 and SLC14A1, including 58 predicted frameshifts. Software predictions were validated by serology in 108 participants; one case in the FY blood group and three cases in the JK blood group were discrepant. Investigation revealed that these discrepancies resulted from (1) clerical error, (2) serologic failure to detect weak antigenic expression and (3) a frameshift variant absent in blood group databases. CONCLUSION: DTM-Tools can be employed for rapid Kell, Duffy and Kidd blood group antigen prediction from existing ES data sets; for discrepancies detected in the validation data set, software predictions proved accurate. DTM-Tools is open-source and in continuous development.

Comparative effectiveness research in critically ill patients: risks associated with mischaracterising usual care.

Comparative effectiveness research can help guide the use of common, routine medical practices. However, to be safe and informative, such trials must include at least one treatment arm that accurately portrays current practices. While comparative effectiveness research is widely perceived as safe and to involve no or only minimal risks, these assumptions may not hold true if unrecognised deviations from usual care exist in one or more study arms. For critically ill subjects in particular, such practice deviations may increase the risk of death or injury and undermine safety monitoring. Furthermore, unrecognised unusual care seems likely to corrupt informed consent documents, with underappreciated risks shrouded under the reassuring "comparative effectiveness" research label. At present, oversight measures are inadequate to ensure that research subjects enrolled in comparative effectiveness trials are actually receiving usual and not unusual care. Oversight by governmental and non-governmental entities with appropriate expertise, empowered to ensure that current clinical practice has been properly represented, could help prevent occurrences in clinical trials of unusual care masquerading as usual care.

In canine bacterial pneumonia circulating granulocyte counts determine outcome from donor cells.

BACKGROUND: In experimental canine septic shock, depressed circulating granulocyte counts were associated with a poor outcome and increasing counts with prophylactic granulocyte colony-stimulating factor (G-CSF) improved outcome. Therapeutic G-CSF, in contrast, did not improve circulating counts or outcome, and therefore investigation was undertaken to determine whether transfusing granulocytes therapeutically would improve outcome. STUDY DESIGN AND METHODS: Twenty-eight purpose-bred beagles underwent an intrabronchial Staphylococcus aureus challenge and 4 hours later were randomly assigned to granulocyte (40-100 × 109 cells) or plasma transfusion. RESULTS: Granulocyte transfusion significantly expanded the low circulating counts for hours compared to septic controls but was not associated with significant mortality benefit (1/14, 7% vs. 2/14, 14%, respectively; p = 0.29). Septic animals with higher granulocyte count at 4 hours (median [interquartile range] of 3.81 3.39-5.05] vs. 1.77 [1.25-2.50]) had significantly increased survival independent of whether they were transfused with granulocytes. In a subgroup analysis, animals with higher circulating granulocyte counts receiving donor granulocytes had worsened lung injury compared to septic controls. Conversely, donor granulocytes decreased lung injury in septic animals with lower counts. CONCLUSION: During bacterial pneumonia, circulating counts predict the outcome of transfusing granulocytes. With low but normal counts, transfusing granulocytes does not improve survival and injures the lung, whereas for animals with very low counts, but not absolute neutropenia, granulocyte transfusion improves lung function.

RBC Storage Lesion Studies in Humans and Experimental Models of Shock.

The finding of toxicity in a meta-analysis of observational clinical studies of transfused longer stored red blood cells (RBC) and ethical issues surrounding aging blood for human studies prompted us to develop an experimental model of RBC transfusion. Transfusing older RBCs during canine pneumonia increased mortality rates. Toxicity was associated with in vivo hemolysis with release of cell-free hemoglobin (CFH) and iron. CFH can scavenge nitric oxide, causing vasoconstriction and endothelial injury. Iron, an essential bacterial nutrient, can worsen infections. This toxicity was seen at commonly transfused blood volumes (2 units) and was altered by the severity of pneumonia. Washing longer-stored RBCs mitigated these detrimental effects, but washing fresh RBCs actually increased them. In contrast to septic shock, transfused longer stored RBCs proved beneficial in hemorrhagic shock by decreasing reperfusion injury. Intravenous iron was equivalent in toxicity to transfusion of longer stored RBCs and both should be avoided during infection. Storage of longer-stored RBCs at 2 °C instead of higher standard temperatures (4-6 °C) minimized the release of CFH and iron. Haptoglobin, a plasma protein that binds CFH and increases its clearance, minimizes the toxic effects of longer-stored RBCs during infection and is a biologically plausible novel approach to treat septic shock.

Haptoglobin therapy has differential effects depending on severity of canine septic shock and cell-free hemoglobin level.

BACKGROUND: During sepsis, higher plasma cell-free hemoglobin (CFH) levels portend worse outcomes. In sepsis models, plasma proteins that bind CFH improve survival. In our canine antibiotic-treated Staphylococcus aureus pneumonia model, with and without red blood cell (RBC) exchange transfusion, commercial human haptoglobin (Hp) concentrates bound and compartmentalized CFH intravascularly, increased CFH clearance, and lowered iron levels, improving shock, lung injury, and survival. We now investigate in our model how very high CFH levels and treatment time affect Hp's beneficial effects. MATERIALS AND METHODS: Two separate canine pneumonia sepsis Hp studies were undertaken: one with exchange transfusion of RBCs after prolonged storage to raise CFH to very high levels and another with rapidly lethal sepsis alone to shorten time to treat. All animals received continuous standard intensive care unit supportive care for 96 hours. RESULTS: Older RBCs markedly elevated plasma CFH levels and, when combined with Hp therapy, created supraphysiologic CFH-Hp complexes that did not increase CFH or iron clearance or improve lung injury and survival. In a rapidly lethal bacterial challenge model without RBC transfusion, Hp binding did not increase clearance of complexes or iron or show benefits seen previously in the less lethal model. DISCUSSION: High-level CFH-Hp complexes may impair clearance mechanisms and eliminate Hp's beneficial effect during sepsis. Rapidly lethal sepsis narrows the therapeutic window for CFH and iron clearance, also decreasing Hp's beneficial effects. In designing clinical trials, dosing and kinetics may be critical factors if Hp infusion is used to treat sepsis.

Impact of different standard red blood cell storage temperatures on human and canine RBC hemolysis and chromium survival.

BACKGROUND: Storage temperature is a critical factor for maintaining red-blood cell (RBC) viability, especially during prolonged cold storage. The target range of 1 to 6°C was established decades ago and may no longer be optimal for current blood-banking practices. STUDY DESIGN AND METHODS: Human and canine RBCs were collected under standard conditions and stored in precision-controlled refrigerators at 2°C, 4°C, or 6°C. RESULTS: During 42-day storage, human and canine RBCs showed progressive increases in supernatant non-transferrin-bound iron, cell-free hemoglobin, base deficit, and lactate levels that were overall greater at 6°C and 4°C than at 2°C. Animals transfused with 7-day-old RBCs had similar plasma cell-free hemoglobin and non-transferrin-bound iron levels at 1 to 72 hours for all three temperature conditions by chromium-51 recovery analysis. However, animals transfused with 35-day-old RBCs stored at higher temperatures developed plasma elevations in non-transferrin-bound iron and cell-free hemoglobin at 24 and 72 hours. Despite apparent impaired 35-day storage at 4°C and 6°C compared to 2°C, posttransfusion chromium-51 recovery at 24 hours was superior at higher temperatures. This finding was confounded by a preparation artifact related to an interaction between temperature and storage duration that leads to removal of fragile cells with repeated washing of the radiolabeled RBC test sample and renders the test sample unrepresentative of the stored unit. CONCLUSIONS: RBCs stored at the lower bounds of the temperature range are less metabolically active and produce less anaerobic acidosis and hemolysis, leading to a more suitable transfusion product. The higher refrigeration temperatures are not optimal during extended RBC storage and may confound chromium viability studies.

Transfusion support for matched sibling allogeneic hematopoietic stem cell transplantation (1993-2010): factors that predict intensity and time to transfusion independence.

BACKGROUND: Patients undergoing allogeneic hematopoietic stem cell transplant require variable, often extensive transfusion support. Identification of factors that predict urgent, intensive, or special needs should improve management of these patients. STUDY DESIGN AND METHODS: This is a retrospective study of red blood cell (RBC) and platelet transfusion support provided for sequential matched sibling donor allogeneic transplants conducted at the Clinical Center, National Institutes of Health, from 1993 through 2010. Factors potentially important for predicting quantity of RBC and platelet transfusions, and time to transfusion independence through Day 200 following hematopoietic stem cell transplantation were evaluated. RESULTS: Subjects (n = 800) received 10,591 RBC and 10,199 platelet transfusions. Multivariable analysis demonstrated that the need for RBC pretransplant, CD34+ dose, transplant year, diagnostic category, and ABO match were significantly independently associated with quantity of RBC transfusions during Days 0 through 30. Only pretransplant need for RBCs, CD34+ dose, and transplant year had significance during Days 0 through 100. Similar analyses for quantity of platelet transfusions demonstrated that for both Days 0 through 30 and 0 through 100 significant factors were need for platelet support before transplant, CD34+ dose, transplant year, and transplant regimen. Of note, long term, during Days 101 through 200, only CD34+ dose remained significant for quantity of RBC and platelet transfusions. Analysis of time to transfusion independence demonstrated that patients with ABO major mismatches required longer to achieve freedom from RBC transfusion support compared to identical matches or those with minor mismatches. CONCLUSION: Patient-specific factors including CD34+ dose and ABO match of the graft should be given particular consideration by transfusion services when planning support of patients receiving allogeneic hematopoietic stem cell transplant.

Genomic coordinates and continental distribution of 120 blood group variants reported by the 1000 Genomes Project.

BACKGROUND: The 1000 Genomes Project provides a database of genomic variants from whole genome sequencing of 2504 individuals across five continental superpopulations. This database can enrich our background knowledge of worldwide blood group variant geographic distribution and identify novel variants of potential clinical significance. STUDY DESIGN AND METHODS: The 1000 Genomes database was analyzed to 1) expand knowledge about continental distributions of known blood group variants, 2) identify novel variants with antigenic potential and their geographic association, and 3) establish a baseline scaffold of chromosomal coordinates to translate next-generation sequencing output files into a predicted red blood cell (RBC) phenotype. RESULTS: Forty-two genes were investigated. A total of 604 known variants were mapped to the GRCh37 assembly; 120 of these were reported by 1000 Genomes in at least one superpopulation. All queried variants, including the ACKR1 promoter silencing mutation, are located within exon pull-down boundaries. The analysis yielded 41 novel population distributions for 34 known variants, as well as 12 novel blood group variants that warrant further validation and study. Four prediction algorithms collectively flagged 79 of 109 (72%) known antigenic or enzymatically detrimental blood group variants, while 4 of 12 variants that do not result in an altered RBC phenotype were flagged as deleterious. CONCLUSION: Next-generation sequencing has known potential for high-throughput and extended RBC phenotype prediction; a database of GRCh37 and GRCh38 chromosomal coordinates for 120 worldwide blood group variants is provided as a basis for this clinical application.

Haptoglobin improves shock, lung injury, and survival in canine pneumonia.

During the last half-century, numerous antiinflammatory agents were tested in dozens of clinical trials and have proven ineffective for treating septic shock. The observation in multiple studies that cell-free hemoglobin (CFH) levels are elevated during clinical sepsis and that the degree of increase correlates with higher mortality suggests an alternative approach. Human haptoglobin binds CFH with high affinity and, therefore, can potentially reduce iron availability and oxidative activity. CFH levels are elevated over approximately 24-48 hours in our antibiotic-treated canine model of S. aureus pneumonia that simulates the cardiovascular abnormalities of human septic shock. In this 96-hour model, resuscitative treatments, mechanical ventilation, sedation, and continuous care are translatable to management in human intensive care units. We found, in this S. aureus pneumonia model inducing septic shock, that commercial human haptoglobin concentrate infusions over 48-hours bind canine CFH, increase CFH clearance, and lower circulating iron. Over the 96-hour study, this treatment was associated with an improved metabolic profile (pH, lactate), less lung injury, reversal of shock, and increased survival. Haptoglobin binding compartmentalized CFH to the intravascular space. This observation, in combination with increasing CFHs clearance, reduced available iron as a potential source of bacterial nutrition while decreasing the ability for CFH and iron to cause extravascular oxidative tissue injury. In contrast, haptoglobin therapy had no measurable antiinflammatory effect on elevations in proinflammatory C-reactive protein and cytokine levels. Haptoglobin therapy enhances normal host defense mechanisms in contrast to previously studied antiinflammatory sepsis therapies, making it a biologically plausible novel approach to treat septic shock.