Red Cell Transfusion and O2 delivery – Do the Math

James A. DiNardo MD and David Faraoni MD

Over the past couple of weeks, we’ve discussed the issue of perioperative red blood cell transfusions extensively in the PAAD.  Should we target a specific hemoglobin level or the effects of tissue ischemia if oxygen delivery does not match cellular demand?  As many of you know, Dr. Jim DiNardo is a frequent contributor to the PAAD and is a member of the PAAD’s cardiac anesthesia review team.  He together with Dr. David Faraoni recently published a paper on this topic that I thought would help define and illuminate many of these issues and I asked them to share their thoughts with our readership.  Myron Yaster MD

PS: As we enter into the holiday season, we would like to ask all of you, our reader/subscribers, to offer “what you are grateful for”?  Send your responses to Myron (myasterster@gmail.com) and we will publish them on Christmas eve. 

Editorial

Faraoni D, DiNardo JA. Commentary: Red blood cells transfusion in patients undergoing congenital cardiac surgery: Still far from physiology-based practice. J Thorac Cardiovasc Surg. 2022 Oct 27:S0022-5223(22)01147-3. doi: 10.1016/j.jtcvs.2022.10.025. Online ahead of print.PMID: 36404142

Due to a lack of evidence, red blood cell (RBC) transfusions are routinely administered based on clinical judgments without evidence that oxygen consumption (VO2) has become oxygen delivery (DO2) dependent and potentially limited by hemoglobin concentration 1. A recent PAAD summarized an excellent, comprehensive review article 2 addressing the complicated issue of defining erythrocyte transfusion thresholds in children with an emphasis on incorporation of critical hemoglobin (Hb) thresholds and physiologic endpoints in decision making. Use of meaningful physiologic endpoints to trigger RBC transfusion is long overdue and is a direct challenge to attempts to define hemoglobin thresholds based on expert consensus.

Using a modified mathematical model of DO2 in single ventricle physiology, Ahmed et al. 3 challenge the assertion, put forward by a pediatric expertise initiative 4, that transfusion beyond a Hb of 9 g/dL for patients with cyanotic heart disease should be avoided. The authors point out that the expertise initiative acknowledged that this was a weak recommendation based on low-quality pediatric evidence and in fact, recommended that transfusion be based on measurable clinical indices such as arterial oxygen saturation [SaO2], systemic venous oxygen saturation [SsvO2], lactate, and clinical signs and symptoms.

The model is used to analyze the effects of varying Hb and Qp:Qs on DO2, SaO2, SsvO2, and oxygen extraction ratio (OER) defined as (SaO2 -SsvO2)/SaO2. The authors reasonably fix total cardiac output Qt = Qp + Qs at 6 L/m2/min, pulmonary vein oxygen saturation (SpvO2) at 97%, and total body oxygen consumption (CVO2) at 150 mL/ m2/min. In single ventricle physiology SaO2 and SsvO2 are inextricably linked because SaO2 is determined by the weighted averages of SPVO2 and SsvO2 flows and SsvO2 is determined by CVO2 for a given SaO2. SsvO2 ≤ 30% inevitably leads to lactate production indicative of cell ischemia and death. In fact, the authors demonstrate that given the models assumptions, to operate above the critical oxygen economy boundary (SsvO2 » 40%) and maintain SaO2 >70% at a Qp:Qs » 1.0 would require either increasing the cardiac index to » 9 L/m2/min or increasing the hemoglobin to greater than 13 g/dL. Obviously, in a volume overloaded ventricle an increase in Qt from 6 L/m2/min to 9 L/m2/min (a 50% increase) may not be achievable even with substantial inotropic and vasodilator therapy. Alternatively stated, with a Hb of 9 g/dL a patient with this physiology would be unable to tolerate even minor decreases in SpvO2 or Qt or increases in CVO2 without anaerobic metabolism ensuing. This is an entirely unrealistic expectation given the normal stresses of crying, eating, or developing a fever. Finally, the greatest improvement in arterial and venous saturation arises when Hb is augmented from levels below 12 g/dL.

While elegant, the mathematical model only considers systemic DO2. As demonstrated in several experiments, increases in systemic DO2 does not necessarily result in a proportional increases in microcirculatory DO2 at the level of the end organs and tissues. Using a mathematical model of the microcirculation, Li et al. found that the influence of RBC transfusion on DO2 varies significantly depending on the ability of the vascular endothelium to respond to increases in wall shear stress induced by blood viscosity 5. In a microvasculature responsive to wall shear stress, RBC transfusion increases nitric oxide production and induces vasodilation, which results in increased DO2 despite the increase in blood viscosity. In the presence of nonresponsive endothelium, RBC transfusion may lower DO2 due to an unopposed increase in blood viscosity. Furthermore, a minimum viscosity level is necessary to generate the shear stress and the release of nitric oxide and prostacyclin needed to maximize cardiac output. Similarly, Zimmerman and colleagues have suggested that the optimal hematocrit level should be defined as the hematocrit at which increased oxygen-carrying capacity and oxygen-diffusive losses due to reduced blood flow velocity from increased viscosity are balanced such that DO2 is maximized 6.

 A recent study by Savorgnan and colleagues provide clinical evidence that the effect of RBC transfusion on the increase in DO2 at the microvasculature level of the coronary artery circulation, is extremely complex. 7 The authors captured and analyzed high-frequency physiologic data 30 minutes prior to and during the 120 minutes of RBC transfusion in 73 neonates with hypoplastic left heart syndrome following the Norwood procedure. They found a significant increase in the ST segment variability and evidence of myocardial ischemia  temporally  associated  with  RBC transfusions. While the mechanism of these observations is unclear, they reinforce the concept that systemic DO2 is not necessarily reflective of microvascular DO2.

Our increased understanding of the universe has paralleled development of tools which allow us to see beyond what is visible to the naked eye. Likewise, an increased understanding of microvascular and tissue DO2 will require us to acquire the tools necessary to see beyond systemic DO2. In the meantime, a single transfusion threshold is not and will never be a viable approach to optimizing oxygen delivery.

References

1.          Faraoni D, DiNardo JA. Commentary: Red blood cells transfusion in patients undergoing congenital cardiac surgery: Still far from physiology-based practice. J Thorac Cardiovasc Surg 2022:S0022522322011473.

2.          Downey LA, Goobie SM. Perioperative Pediatric Erythrocyte Transfusions: Incorporating Hemoglobin Thresholds and Physiologic Parameters in Decision-making. Anesthesiology 2022;137:604–19.

3.          Ahmed M, Acosta SI, Hoffman GM, Tweddell JS, Ghanayem NS. Mathematical analysis of hemoglobin target in univentricular parallel circulation. J Thorac Cardiovasc Surg 2022:S0022-5223(22)01036-4.

4.          Valentine SL, Bembea MM, Muszynski JA, Cholette JM, Doctor A, Spinella PC, Steiner ME, Tucci M, Hassan NE, Parker RI, Lacroix J, Argent A, Carson JL, Remy KE, Demaret P, Emeriaud G, Kneyber MCJ, Guzzetta N, Hall MW, Macrae D, Karam O, Russell RT, Stricker PA, Vogel AM, Tasker RC, Turgeon AF, Schwartz SM, Willems A, Josephson CD, Luban NLC, et al. Consensus Recommendations for RBC Transfusion Practice in Critically Ill Children From the Pediatric Critical Care Transfusion and Anemia Expertise Initiative. Pediatr Crit Care Med J Soc Crit Care Med World Fed Pediatr Intensive Crit Care Soc 2018;19:884–98.

5.          Li W, Tsai AG, Intaglietta M, Tartakovsky DM. A model of anemic tissue perfusion after blood transfusion shows critical role of endothelial response to shear stress stimuli. J Appl Physiol 1985 2021;131:1815–23.

6.          Zimmerman RA, Tsai AG, Intaglietta M, Tartakovsky DM. A Mechanistic Analysis of Possible Blood Transfusion Failure to Increase Circulatory Oxygen Delivery in Anemic Patients. Ann Biomed Eng 2019;47:1094–105.

7.          Savorgnan F, Bhat PN, Checchia PA, Acosta S, Tume SC, Lasa JJ, Asadourian V, Achuff B-J, Flores S, Ahmed M, Crouthamel DI, Loomba RS, Bronicki RA. RBC Transfusion Induced ST Segment Variability Following the Norwood Procedure. Crit Care Explor 2021;3:e0417.