See Article, p 1077
Pediatric cardiac patients undergoing cardiac surgery are frequently exposed to red blood cell (RBC) transfusions. Transfusions are usually added into the cardiopulmonary bypass (CPB) to avoid prime-induced hemodilution and are commonly administered after CPB in the context of excessive blood loss or to optimize hemoglobin delivery to vital organs. Although studies have shown that liberal transfusions are not superior to restrictive transfusions in children with cyanotic and noncyanotic heart diseases,1,2 liberal transfusion strategies are commonly used in the perioperative period of pediatric cardiac surgery despite the published recommendations.3,4
In this issue of Anesthesia & Analgesia, Long et al5 published the results of an analysis of the Society of Thoracic Surgeons (STS) and Congenital Cardiac Anesthesia Society (CCAS) database in which the association between postcardiac surgery hematocrit values and postoperative complications or mortality in pediatric cardiac surgical patients was examined. The authors performed a retrospective analysis of data included in the STS-CHSD (Society of Thoracic Surgeons Congenital Heart Surgery Database) between 2014 and 2019. They performed multivariable logistic regressions to assess the relationship between postoperative hematocrit (defined as hematocrit at admission to intensive care) and mortality or any major complication after adjustment for covariates included in the STS-CHSD mortality risk model. The definition of cyanotic versus noncyanotic lesions was an expected postsurgical arterial oxygen saturation of <92% and ≥92%, respectively. The study included 27,462 operations in 4909 (17.9%) cyanotic and 22,553 (82.1%) noncyanotic patients. Each 5% increase in postoperative hematocrit over 42% for cyanotic patients and 38% for noncyanotic patients was associated with a significant increase in the odds of operative mortality and major complications. The authors concluded that since post-CPB hematocrits are controllable, opportunities exist to better define targets for transfusion and avoid excessive administration of RBCs. However, the authors admit that the motivation for providers to transfuse to high post-CPB hematocrits and the optimal target for post-CPB hematocrit remain unclear.
The association between RBC transfusion and poor outcomes has been reported and debated for decades. Despite the publication of a large number of clinical trials comparing transfusion triggers and showing the absence of benefit associated with higher hemoglobin targets6 and an even larger number of retrospective studies associating transfusion with higher incidences of complications, clinicians remain skeptical in adopting restrictive transfusion strategies in their clinical practice. Let us try to understand the motivation for this skepticism.
First, the quality of the published literature is debatable as both prospective randomized trials and retrospective studies have some important limitations. Perhaps most important is the fact that the study cohorts are not necessarily reflective of the patient population cared for by the clinician responsible for transfusion decisions.7 Prospective randomized trials offer the benefit of controlling for an intervention (eg, transfusion), while the overall patient’s management is standardized. However, these studies usually exclude a large proportion of pediatric cardiac surgical patients because of (1) the challenge of recruiting pediatric patients in clinical trials and (2) the exclusion criteria required to make the inclusion of patients in a clinical trial ethical.8,9 Although those prospective transfusion studies are very important, it is often difficult to change clinical practice based on their results since our patient population extends beyond the patients included in those studies. Retrospective analysis of large datasets offers the advantage of allowing the researchers to include a large number of patients and is thought to better reflect what happens in “real life.” However, despite all the statistical methods available to adjust for confounding factors, the results of retrospective analyses must always be interpreted with caution due to high likelihood that they can only identify associations and cannot be used to prove that the treatment (eg, transfusion) is the cause of the association. The study by Long et al is a great example of why association does not guarantee causation. Although the authors report an association between higher hematocrit and poor outcomes, it is impossible to prove that RBC transfusion is to blame for the association. Despite their adjusted analysis, it is highly possible that variables for which there was no adjustment (eg, bleeding, hypoxemia, and low cardiac output) lead to transfusion and contributed to the poor outcome observed.
The goal of RBC transfusion is to restore or maintain a target hemoglobin level identified by the physician as necessary to maintain oxygen delivery (DO2) matched to oxygen consumption (VO2) to allow for adequate organ perfusion and to avoid utilization of anaerobic metabolism by tissues. Although this physiologic concept has long been appreciated, its application at the bedside is sometimes challenging. It is indeed well accepted that the relationship between DO2 and VO2, and determination of the critical DO2 or critical hemoglobin will vary depending on the clinical scenario. For example, in the presence of high metabolic demand (eg, sepsis), it has been demonstrated that the critical hemoglobin required to maintain aerobic metabolism is shifted toward higher critical hemoglobin.10 On the contrary, general anesthesia decreases VO2 and, therefore, leads to a shift of the critical hemoglobin toward the lower left of the relationship.11 It is therefore easily understandable that the hemoglobin that needs to be targeted to optimize DO2 and VO2 is a moving target that is highly variable depending on the clinical situation. In addition to that, the pediatric cardiac surgical population includes a significant proportion of patients who suffer from chronic hypoxemia and who require higher hemoglobin to provide oxygen to the organs and tissues. Although we do agree with the authors that it is important to target the appropriate hemoglobin level, the definition of the optimal target in the perioperative period wherein fluctuations in DO2 and VO2 are the norm is challenging.
Finally, transfusion may be bad for our pediatric cardiac patients unless they are bleeding.12 The administration of RBC in the presence of excessive bleeding is crucial to avoid bleeding-induced hemodilution and anemia. In the context of massive bleeding, it is not rare to find the patients with hemoglobin higher than what the clinicians initially targeted. Although an association between hemoglobin and poor outcome was found in this study, the association between bleeding and poor outcome has been well described.13,14 Because massive transfusion will never be administered in the absence of bleeding, identifying the cause of poor outcomes between transfusion and bleeding has proven to be very challenging or impossible.
In summary, the study published by Long et al in this issue of Anesthesia & Analgesia provides more evidence that there exists a potential association between high hemoglobin and poor outcome in pediatric cardiac surgical patients. Whether transfusion or the indications for the transfusion or the combination of both led to poor outcome cannot be determined. In the meantime, it is important to remind clinicians caring for pediatric cardiac patients that the optimal hemoglobin is highly variable between the patients and is a moving target over time. Because a single transfusion threshold for all patients will never be a viable approach, clinical tools to rapidly and reliably determine DO2 and VO2 so that transfusion of RBC can be evidence-based are urgently needed.
Name: David Faraoni, MD, PhD, FAHA.
Contribution: This author helped write the manuscript.
Name: James A. DiNardo, MD, FAAP.
Contribution: This author helped write the manuscript.
This manuscript was handled by: Susan Goobie, MD, FRCPC.
1. Willems A, Harrington K, Lacroix J, et al.; TRIPICU investigators; Canadian Critical Care Trials Group; Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network. Comparison of two red-cell transfusion strategies after pediatric cardiac surgery: a subgroup analysis. Crit Care Med. 2010;38:649–656.
2. Cholette JM, Rubenstein JS, Alfieris GM, Powers KS, Eaton M, Lerner NB. Children with single-ventricle physiology do not benefit from higher hemoglobin levels post cavopulmonary connection: results of a prospective, randomized, controlled trial of a restrictive versus liberal red-cell transfusion strategy. Pediatr Crit Care Med. 2011;12:39–45.
3. Cholette JM, Willems A, Valentine SL, Bateman ST, Schwartz SM; Pediatric Critical Care Transfusion and Anemia Expertise Initiative (TAXI); Pediatric Critical Care Blood Research Network (BloodNet), and the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network. Recommendations on RBC transfusion in infants and children with acquired and congenital heart disease from the pediatric critical care transfusion and anemia expertise initiative. Pediatr Crit Care Med. 2018;19:S137–S148.
4. Faraoni D, Meier J, New HV, Van der Linden PJ, Hunt BJ. Patient blood management for neonates and children undergoing cardiac surgery: 2019 NATA guidelines. J Cardiothorac Vasc Anesth. 2019;33:3249–3263.
5. Long JB, Engorn BM, Hill KD, et al. Postoperative hematocrit and adverse outcomes in pediatric cardiac surgery patients: a cross-sectional study from the Society of Thoracic Surgeons and Congential Cardiac Anesthesia Society database collaboration. Anesth Analg. 2021;133:1077–1088.
6. Kashani HH, Lodewyks C, Kavosh MS, et al. The effect of restrictive versus liberal transfusion strategies on longer-term outcomes after cardiac surgery: a systematic review and meta-analysis with trial sequential analysis. Can J Anaesth. 2020;67:577–587.
7. Faraoni D, Schaefer ST. Randomized controlled trials vs. observational studies: why not just live together? BMC Anesthesiol. 2016;16:102.
8. Trentino KM, Farmer SL, Isbister JP, et al. Restrictive versus liberal transfusion trials: are they asking the right question? Anesth Analg. 2020;131:1950–1955.
9. Trentino KM, Farmer SL, Leahy MF, et al. Systematic reviews and meta-analyses comparing mortality in restrictive and liberal haemoglobin thresholds for red cell transfusion: an overview of systematic reviews. BMC Med. 2020;18:154.
10. Spec-Marn A, Tos L, Kremzar B, Milic-Emili J, Ranieri VM. Oxygen delivery-consumption relationship in adult respiratory distress syndrome patients: the effects of sepsis. J Crit Care. 1993;8:43–50.
11. Van der Linden P, Gilbart E, Engelman E, Schmartz D, Vincent JL. Effects of anesthetic agents on systemic critical O2 delivery. J Appl Physiol (1985). 1991;71:83–93.
12. DiNardo JA. Blood transfusions might be bad for you; that is unless you are bleeding. Anesth Analg. 2013;116:1201–1203.
13. Goobie SM, DiNardo JA, Faraoni D. Relationship between transfusion volume and outcomes in children undergoing noncardiac surgery. Transfusion. 2016;56:2487–2494.
14. Faraoni D, Emani S, Halpin E, et al. Relationship between transfusion of blood products and the incidence of thrombotic complications in neonates and infants undergoing cardiac surgery. J Cardiothorac Vasc Anesth. 2017;31:1943–1948.