Known and Unknown Unknowns in Making Erythrocyte Transfusion Decisions

Erythrocyte transfusion, used as both a prophylactic and therapeutic intervention, is cellular transplantation that comes with consequent sequelae including immediate1–3 long-term4 engraftment of donor leukocytes in the recipient, other immunologic adversities. It is, therefore, incumbent upon clinicians to identify when erythrocyte transfusion indicated. This issue Anesthesiology contains an important publication by Zeroual et al.5 investigates consequence increasing restrictiveness guidelines.Non–actively bleeding postcardiac surgery patients intensive care unit (ICU) whom hemoglobin concentration fell less than 9 g/dl were randomly allocated al. either standard-of-care or experimental arm which individuals would be transfused only if their was superior vena cava oxyhemoglobin saturation (which does not include contribution from inferior cava, thus distinguishing it true mixed venous saturation, usually clinically obtained pulmonary artery) equal 65%. The addition this criterion for naturally led decreased incidence ICU (from 100% 68%), half who untransfused remaining at hospital discharge; fraction requiring randomization discharge also significantly different between two arms (P = 0.0058).The importance al.’s work implementation adding physiologic criterion, sole traditional concentration. noted earlier article6 understand oxygen delivery no longer meets need, accurate measure oxygenation tissue level required, along understanding physiologic/pathophysiologic consequences anemia critical organ(s) each individual patient. Two decades later, decision-making still requires: (1) ability define physiologically required; (2) clear relative safety transfusing erythrocytes versus (untransfused anemia).Zeroual believe can need. All surrogates, are second best real interest, must validated, should latter cannot assessed. Unfortunately, circumstance we have been offered data support notion 65% inadequate: systemic organ measures inadequate presented. Decreased content increased extraction—a normal response—but demonstration response mechanisms (increased cardiac output, extraction, both) sufficient compensate lesser Healthy humans respond acute severe output extraction measured (pulmonary (the 5 mean 69.6%), but without evidence (normal, unchanged consumption lactate concentration).7 However, all organs equally sensitive decrements delivery, brain likely being most sensitive. central processing8 subtle cognitive function deficits 6 g/dl, (and more so g/dl),9 despite absence markers delivery. These reversed augmentation transfusion9,10 breathing oxygen11 applied approximately 30 min after onset anemia. unknown whether anemia-induced fully reversible duration. Subtle generally detectible operating room ICU. Thus, seek direct validated surrogates assess function. While assessment might available few selected patients, these typical one units. Nevertheless, point field correct direction: finding implementing criteria dictate transfusion.The trial neither designed nor powered safety, there assessing higher nervous system results ischemic events show tantalizing four-fold, statistically insignificant, numerical difference: 2% standard group 8% group; 95% CI corresponding odds ratio ranges 0.4 214, reflecting overwhelming uncertainty need substantially larger sample size. Renal appropriately assessed using Kidney Disease Improving Global Outcomes criteria, again, inconclusive 0.28 1.48. helpful such assessments compare pretherapy posttherapy data, well biomarkers (e.g., urinary N-acetyl-β-D-glucosaminidase neutrophil gelatinase–associated lipocalin). As events, renal deterioration stage 3 evaluated study.Evidence guiding practice has times come misleading retrospective analysis observational databases, suggest outcomes, provide definitive results. Randomized trials testing hypotheses generated analyses lead considerable expenditure funds, personnel, time could better spent otherwise.12 For example, problematic patient investigating efficacy stored 2 weeks weeks13 spawned many prospective clinical trials, showed difference two.14–20 In smaller studies healthy humans, previously shown infusion autologous fresh did differ reverse deficits10 effects on gas exchange.21 confirmed small randomized patients.22Appropriately designed, executed, analyzed yield rigorous means hypotheses. they population-based intervention effects, do give insights into how treat specific patients. fields medicine, anesthesiologists, intensivists, specialists heterogeneous population presenting own pathophysiology may make them susceptible treatment courses. clinician able those subpopulations apply. context, evaluation patient’s pathophysiological our appropriate decision transfusion. examined restrictive (8 g/dl) liberal (10 trigger high-risk undergoing hip fracture surgical repair found mortality groups. 14.1% 4.8% due cardiovascular symptoms referable < 0.00001), rather having reaching assigned trigger. issues difficult comprehend individualization debated,23–25 some arguing against concept.26 Our inability evaluate adequacy oxygenation, prevent, treat, unacceptable patient, genesis range recommended American Society Anesthesiologists.27,28Waiting ill-advised, detection, caused irreversible damage. Accordingly, consequences. To end, organs, highly associated, changes parameters sought preventative intervention. Only transfuse based immediate imminent will benefit risk transfusion.Dr. Weiskopf consulted National Institutes Health (Bethesda, Maryland), U.S. Food Drug Administration (Washington, D.C.), Department Defense D.C.) regarding programs. He sponsors hemoglobin-based carriers, received any compensation commercial entities past yr. Dr. Cook consults TerumoBCT (Lakewood, Colorado).