Secondary Logo

Journal Logo

Pediatric Anesthesiology: Original Clinical Research Report

Red Blood Cell Transfusion in Pediatric Orthotopic Liver Transplantation: What a Difference a Few Decades Make

Tran, Lieu T. MD*; Mazariegos, George V. MD; Damian, Daniela MD; Davis, Peter J. MD

Author Information
doi: 10.1213/ANE.0000000000003832
  • Free



  • Question: How do current rates of red blood cell transfusion for pediatric orthotopic liver transplantations compare to previous reports?
  • Findings: Evaluation of current transfusion practices at our institution reveals that most pediatric patients undergoing liver transplantation receive <1 blood volume of packed red blood cells, with >1 in 4 transplantations requiring no transfusion at all.
  • Meaning: In contrast to historical reports, current management of liver transplantation at our institution often is either transfusion free or involves markedly smaller volume transfusions, reflecting notable advancements in the field over the past few decades.

Liver transplantation in children is often associated with coagulopathy and significant blood loss. In 1985, Borland et al1 reported the anesthetic management of 68 liver transplants from our institution, the Children’s Hospital of Pittsburgh of University of Pittsburgh Medical Center (UPMC).1 In addition to reporting patient demographics and intraoperative management, Borland et al1 also reported on blood loss, an average of 5.4 blood volumes (BVs) and a range of 0.5–25 BV. Blood loss was the highest among patients with biliary atresia, and mortality was reported among 17 of 50 patients within 60 days of surgery. During these early years of liver transplantation, Gartner et al2 from the same institution noted a 2-year overall survival rate of 64%.

Since the articles by Borland et al1 and Gartner et al,2 much has changed in the field of pediatric liver transplantation. In the past few decades, improvements have been made to surgical techniques (piggyback approach, reduced-sized grafts, split livers, and domino livers), along with improvements in organ preservation, anesthetic management, and perioperative medical management. Coupled with these changes have been advancements in our understanding of the pharmacology of immunological agents and the principles governing tolerance and rejection. All these factors have allowed for marked improvements in intraoperative management, and in patient and graft survival. In 2018, Venick et al3 from the University of California, Los Angeles, reported on the outcomes of 1000 pediatric liver transplants performed at a single institution over a 30-year period; they found 1-, 5-, 10-, and 20-year patient survival rates to be 86%, 81%, 78%, and 69%, respectively.

Because much has changed over the last 35 years for pediatric patients undergoing liver transplantation, and because red cell transfusions have been associated with adverse outcomes in some surgical populations, we reviewed our experience regarding red blood cell (RBC) transfusion and compared it to our institution’s previously reported experience with this patient population.


This observational retrospective study was approved by our institutional review board. At the onset of this study, our goal was to assess the trend of RBC transfusion currently and to qualitatively compare this with practices from the past as described by Borland et al.1 Patient data were collected from medical records at the Children’s Hospital of Pittsburgh of the UPMC, a tertiary children’s hospital. All patients who underwent liver transplantation from January 2008 to June 2017 were included. A roster of patients was obtained from our preexisting liver transplant database. Individual charts were reviewed by a single anesthesia provider for demographic data, primary diagnoses, surgical times, RBCs transfused intraoperatively, type of liver transplanted, and mortality. Patients receiving transplantation of a liver in combination with another organ were excluded from the study. Primary and secondary outcomes were volume of RBCs transfused and mortality, respectively.

Information regarding the underlying diagnoses and indications for liver transplantation were obtained from surgical clinic and operative notes. Anesthesia charts were primarily reviewed for transfusion data. When transfusion data were not present in the anesthesia notes, the surgeon’s operative note was reviewed. We reviewed subsequent intensive care unit notes, inpatient floor notes, and outpatient clinic notes for follow-up data that include mortality. Death of a patient is flagged by our electronic medical record system with a color banner.

Estimated BV ranges used for different age groups were as follows: premature neonates, 90 mL/kg; term neonates, 80 mL/kg; infants ≤12 months of age, 75 mL/kg; and children >12 months of age, 70 mL/kg. We used the lower limit of the estimated BV range for age based on Smith’s Anesthesia for Infants and Children.4 During chart review, when RBC transfusion volume was reported as units instead of exact milliliters transfused, 1 packed RBC unit was converted to 350 mL.

We used SPSS Statistics, version 25 (IBM Corp, Armonk, NY, 2017) to obtain descriptive statistics and to generate the figures.

At our institution, one-third of the faculty comprising of 10 pediatric anesthesiologists participate in the pediatric transplant team. Except for a protocol for immunosuppressant therapy administration, there is no standard anesthetic protocol for liver transplantation. We do not have an institutional threshold for RBC transfusion. In general, blood product administration is guided by patient hemodynamics, estimated blood loss as observed on the field, laboratory values, and thromboelastography. Although blood product administration is at the discretion of the anesthesiologist and surgeon, the targeted intraoperative hemoglobin level is 10 g/dL. We do not use routine RBC conservation techniques, and in rare instances will use cell salvage or antifibrinolytic agents.

This manuscript adheres to the applicable Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.


From January 2008 to June 2017, 278 liver transplants were performed at the Children’s Hospital of Pittsburgh of UPMC. The first patient with complete electronic medical record documentation was transplanted on June 23, 2010. Before that, blood transfusion data were collected from either the surgeon’s operative note or anesthesia note, scanned from paper forms into our electronic medical record system. Data for 72 cases were based on scanned paper notes, and data for 206 cases were from electronic medical record entries.

Two hundred fifty-nine primary transplants, 15 second repeat transplants, and 4 third repeat transplants were conducted. Seven patients underwent 2 different liver transplantations during the study period; 271 total patients were studied.

Average age at transplantation was 6.9 years. Of 278 transplant cases, 144 (51.8%) were female (Table). Biliary atresia, maple syrup urine disease, urea cycle defect, and liver mass were the leading indications for transplantation, accounting for 66 (23.7%), 45 (16.2%), 24 (8.6%), and 23 (8.3%) of transplants, respectively. Other indications for transplantation are detailed in the Table.

Patient Characteristics

Seventy-six transplant cases (27.3%) did not require RBC transfusions (Figure 1). Among those transfused, 181 transplant cases (89.6%) required <1 BV. The median BV transfused among all cases was 0.21 (0–9 BV; Q1, 0; Q3, 0.45).

Figure 1.
Figure 1.:
Distribution of blood volumes transfused. There were 278 total liver transplants. The top number indicates number of transplants. The bottom number indicates percent of total transplants.

In the nontransfused group, ages ranged from 7 to 358 months (29.8 years), with the median age of 158 months (13.2 years). Patients with maple syrup urine disease and Crigler–Najjar formed the majority of those who were not transfused, accounting for 37% and 13%, respectively (Figure 2).

Figure 2.
Figure 2.:
Subgroup analysis of transplantations that involved no RBC transfusion. The top number indicates number of transplants. The bottom number indicates the percent out of all the nontransfused cases. AFHF indicates acute fulminant hepatic failure; Alpha-1AT, α-1 antitrypsin deficiency; MSUD, maple syrup urine disease; PSC, primary sclerosing cholangitis.

There was a trend toward higher volume transfusions among infants (median, 0.46 BV) compared to children >12 months of age (0.12 BV; Figure 3). In addition, compared to children who were not transfused any RBCs, those who were transfused were on average much younger. The median age of the nontransfused group was 158 months (13.2 years) compared to 23 months in the transfused group.

Figure 3.
Figure 3.:
Blood volumes transfused in infants versus older children. There were 73 infants (age ≤12 mo) and 205 children (age >12 mo). Box plots are shown with the center of the box as median, the bottom of the box as the 25th percentile, and the top of the box as the 75th percentile. The T bars indicate inner fences, extending to the fifth percentile (lower) and 95th percentile (upper). Individual stars and circles indicate outliers.

By diagnosis, the group requiring the highest median volume transfusions was patients with total parenteral nutrition (TPN)-related liver failure, followed by patients undergoing repeat transplantation (Figure 4). Three of the 4 children with TPN-related liver failure were infants with short gut syndrome; RBC volumes transfused were 10 units (5.2 kg infant), 3 units (5.5 kg infant), and 6.5 units (7.1 kg infant). The fourth patient was a 9 years old with short gut syndrome; she received a smaller volume of RBCs (300 mL) for her weight (30.9 kg) compared to the 3 infants.

Figure 4.
Figure 4.:
Blood volumes transfused by diagnosis. Box plots are shown with the center of the box as median, the bottom of the box as the 25th percentile, and the top of the box as the 75th percentile. The T bars indicate inner fences, extending to the fifth percentile (lower) and 95th percentile (upper). Individual stars and circles indicate outliers. Numbers inside parentheses indicate number of transplants. AFHF indicates acute fulminant hepatic failure; Alpha-1AT, α-1 antitrypsin deficiency; MSUD, maple syrup urine disease; PSC, primary sclerosing cholangitis; TPN, total parenteral nutrition–related liver failure.

Comparison of primary versus repeat transplants shows a trend toward higher volume transfusions with a third transplant (median, 2.71 BV), compared to second transplant (0.43 BV) and primary transplant (0.18 BV).

There were 146 cadaveric reduced, 47 cadaveric whole, 73 living reduced, and 12 living whole liver transplants (domino liver). Living reduced liver transplantations were associated with the highest median blood loss at 0.35 BV (0–9 BV), followed by cadaveric reduced liver with a median of 0.3 BV (0–2.5 BV), followed by cadaveric whole (median, 0.08; 0–6.8 BV) and living whole (domino) (median, 0.08; 0–2.9 BV) liver transplants.

Four of 271 patients (1.5%) died during admission involving liver transplantation. Nine of 271 patients (3.3%) died subsequently. Total mortality was 4.8%. Follow-up mortality data were collected for every patient at the time of this study, which ended on June 14, 2017. Time to follow-up from transplant date ranged from 9 to 3434 days (9.4 years).


Since our institution’s initial publication on pediatric liver transplantation, intraoperative RBC transfusion requirements have changed significantly. Many factors have been responsible for this change. Differences in underlying pathologies, advances in surgical techniques and organ preservation, and improvements in preoperative medical management and intraoperative anesthetic management have all been contributing factors. In addition, a better understanding of the immunological and pharmacological response to organ rejection, coupled with the use of different immunosuppressant agents and protocols, have contributed to improved graft and patient survival.

Even with these advancements, recent studies continue to demonstrate a need for high-volume blood product transfusions in pediatric liver transplant recipients, and transfusion-free liver transplantation remains a challenging goal.5,6 In 2017, Kloesel et al7 from Boston Children’s Hospital reported on a single institution’s experience with 68 pediatric liver transplant patients; all patients required intraoperative RBC administration, and 29 of 68 patients received >1 BV in the perioperative period. Blood transfusions have been shown to be associated with increased patient morbidity and mortality8; in particular to transplantation, transfusions are associated with increased graft failure rates and mortality.9–11

In contrast to our institution’s early experience reported by Borland et al,1 where all patients undergoing liver transplantation received RBCs, 27% of patients in the present study did not receive any RBCs intraoperatively. Among those transfused in our study, 89.6% required <1 BV. Risk factors for higher BV transfusion include younger age, TPN-related liver failure, and repeat transplantation.

Current patient demographics differ markedly compared to those reported by Borland et al.1 Although biliary atresia comprised a large percent of the patients in both series, Borland et al1 noted retransplantation, α-1 antitrypsin deficiency, obstructive disease, and miscellaneous as the leading indications for transplantation. In the current series, biliary atresia, maple syrup urine disease, urea cycle defect, and liver mass were the main indications. Presently, more patients with metabolic diseases are undergoing transplantation. Often, these patients are older and have normal coagulation and hepatic synthetic function.

In the series reported by Borland et al,1 all patients underwent cadaveric whole organ transplantation. Today, more patients undergo living-related, reduced graft, or domino transplants. Although our present experience with reduced-sized grafts suggests a greater transfusion requirement compared to whole liver transplants, children who received living reduced livers were transplanted at a younger age and almost half had biliary atresia.

One of the biggest demographic differences between the former and the present study is the incidence of retransplantation. Retransplantation in the early era was reported as high as 30%,1 whereas our current data suggest 7%. Repeat transplantations were associated with higher volume RBC transfusions. Patients undergoing repeat transplants have adhesions and altered anatomy, resulting in more complex surgeries and longer surgical times, thus accounting for higher volume blood loss.

Within 2 weeks to 2 years of follow-up, Borland et al1 reported a mortality of 34%. In our study, follow-up ranged from 9 days to 9 years, and mortality was at 4.7%.

This study is limited by its retrospective nature. We focused on the use of RBCs, and not other blood products. In addition, we reported on the administration of RBCs as opposed to the report by Borland et al1 on BV loss. Of note, BV loss at the time of the report by Borland et al1 was reflected by the amounts of blood products administered; thus, the comparison may not be exact, but may be similar.

In summary, improvements in surgical techniques and experience coupled with advances in perioperative care have dramatically changed the intraoperative RBC requirements of pediatric patients undergoing liver transplantation over the past few decades. Indications for liver transplantation have broadened. With these changes, both patient and organ survival have improved, with decreased rates of repeat transplantation. Transfusion requirements have markedly diminished, and the likelihood of a transfusion-free procedure has much improved.


Name: Lieu T. Tran, MD.

Contribution: This author helped design the study, collect and interpret the data, write the manuscript, and approve the final manuscript.

Conflicts of Interest: None.

Name: George V. Mazariegos, MD.

Contribution: This author helped design the study, interpret the data, write the manuscript, and approve the final manuscript.

Conflicts of Interest: None.

Name: Daniela Damian, MD.

Contribution: This author helped design the study, interpret the data, write the manuscript, and approve the final manuscript.

Conflicts of Interest: None.

Name: Peter J. Davis, MD.

Contribution: This author helped design the study, interpret the data, write the manuscript, and approve the final manuscript.

Conflicts of Interest: P. J. Davis is a Consultant for Octapharma.

This manuscript was handled by: James A. DiNardo, MD, FAAP.


1. Borland LM, Roule M, Cook DR. Anesthesia for pediatric orthotopic liver transplantation. Anesth Analg. 1985;64:117–124.
2. Gartner JC Jr, Zitelli BJ, Malatack JJ, Shaw BW, Iwatsuki S, Starzl TE. Orthotopic liver transplantation in children: two-year experience with 47 patients. Pediatrics. 1984;74:140–145.
3. Venick RS, Farmer DG, Soto JR, et al. One thousand pediatric liver transplants during thirty years: lessons learned. J Am Coll Surg. 2018;226:355–366.
4. Nemmergut ME, Haile DT, Mauermann WJ. Davis PJ, Cladis FP. Blood conservation and transfusion medicine. In: Smith’s Anesthesia for Infants and Children. 2017:9th ed. Philadelphia, PA: Elsevier414.
5. Fanna M, Baptiste A, Capito C, et al. Preoperative risk factors for intra-operative bleeding in pediatric liver transplantation. Pediatr Transplant. 2016;20:1065–1071.
6. Jin SJ, Kim SK, Choi SS, et al. Risk factors for intraoperative massive transfusion in pediatric liver transplantation: a multivariate analysis. Int J Med Sci. 2017;14:173–180.
7. Kloesel B, Kovatsis PG, Faraoni D, et al. Incidence and predictors of massive bleeding in children undergoing liver transplantation: a single-center retrospective analysis. Paediatr Anaesth. 2017;27:718–725.
8. Kneyber MC, Hersi MI, Twisk JW, Markhorst DG, Plötz FB. Red blood cell transfusion in critically ill children is independently associated with increased mortality. Intensive Care Med. 2007;33:1414–1422.
9. Nacoti M, Cazzaniga S, Lorusso F, et al. The impact of perioperative transfusion of blood products on survival after pediatric liver transplantation. Pediatr Transplant. 2012;16:357–366.
10. Howard-Quijano K, Schwarzenberger JC, Scovotti JC, et al. Increased red blood cell transfusions are associated with worsening outcomes in pediatric heart transplant patients. Anesth Analg. 2013;116:1295–1308.
11. Goobie SM, DiNardo JA, Faraoni D. Relationship between transfusion volume and outcomes in children undergoing noncardiac surgery. Transfusion. 2016;56:2487–2494.
Copyright © 2018 International Anesthesia Research Society