Elsevier Editorial System(tm) for International Journal of Surgery
Title: PREOPERATIVE HAEMATINICS AND TRANSFUSION PROTOCOL REDUCE THE NEED FOR
TRANSFUSION AFTER TOTAL KNEE REPLACEMENT
Article Type: Original Research
Keywords: replacement; allogeneic transfusion; transfusion protocol; anaemia; oral iron.
Corresponding Author: Prof. Manuel Muñoz, PhD
Corresponding Author's Institution:
First Author: Jorge Cuenca, PhD
Order of Authors: Jorge Cuenca, PhD; José A García-Erce, PhD; Fernando Martínez, MD; Rafael Cardona,
MD; Luís Pérez-Serrano, MD; Manuel Muñoz, PhD
Manuscript Region of Origin:
Abstract: Background. Unilateral total knee replacement (TKR) can result in a substantial blood loss and 30-
50% of these patients receive allogeneic blood transfusion (ABT), this transfusion rate may be even higher
among anaemic patients.
Patients and Methods. We assessed the requirements for ABT in 156 consecutive patients undergoing
surgery for primary TKR, who received iron ferrous sulphate (256 mg/day; 80 mg of Fe2+), vitamin C (1000
mg/day) and folic acid (5 mg/day) during the 30-45 days preceding surgery, and who were transfused if Hb
<80 g/L and/or clinical signs/symptoms of acute anaemia/hypoxemia (Group 2). A previous series of 156
TKR patients serves as a control group (Group 1).
Results. Compared to those in the control group, patients in the study group presented a lower transfusion
rate (5.8% vs. 32%, for study and control group, respectively; p<0.01), and a lower transfusion index (1.78 ±
0.44 vs. 2.22 ± 0.65 units per transfused patient, respectively; p<0.05). After patient's stratification according
to a preoperative Hb above or below 130 g/L, the differences in transfusion rate remained significant,
although 19% of patients from group 2 still needed ABT if their preoperative Hb <130 g/L
Conclusion. This protocol seems to be effective for avoiding ABT in non-anaemic TKR patients, whereas for
anaemic patients another blood saving strategy, such us preoperative erythropoietin administration or
postoperative blood salvage, should be added to further increase its effectiveness.
Prof. Michael Baum
International Journal of Surgery
Málaga, January 17, 2006
Dear Prof. Baum:
Please, find enclosed the files corresponding to a new version of our manuscript “PREOPERATIVE
HAEMATINICS AND TRANSFUSION PROTOCOL REDUCE THE NEED FOR TRANSFUSION
AFTER TOTAL KNEE REPLACEMENT” by J. Cuenca, JA. García-Erce, F. Martínez, R. Cardona, L.
Pérez-Serrano, and M. Muñoz, that we should like to be considered for publication in The International
Journal of Surgery.
In this paper we assessed the requirements for allogeneic transfusion in patients undergoing surgery
for primary total knee arthroplasty, who received preoperative hematinics and who were transfused if
Hb <80 g/L and/or clinical signs/symptoms of anemia/hypoxemia. This study was approved by the
Institutional Review Board,
We found that this protocol was easy to implement, well tolerate, and effective for reducing allogenic
transfusion in nonanemic patients. However, it becomes evident that for anemic TKA patients some
additional blood saving method, such us postoperative blood salvage or perioperative erythropoietin
admininstration, should be associated since 20% of them still needs allogeneic transfusion.
All authors have substantially contributed to the design of the study and data discussion, and gave
their approval to the final version of the manuscript. Patients were operated on by the four orthopedics
surgeons, who did the clinical data gathering; blood bank data were provided by JA García-Erce;
statistical analysis was performed by J Cuenca, JA. García-Erce, and M. Muñoz, and manuscript
writing by M. Muñoz. We also declare that no benefits or found were received in support of the study
We look forward to your final decision on our contribution in due time.
Prof. Manuel Muñoz
Manuscript for THE INTERNATIONAL JOURNAL OF SURGERY
Jorge Cuenca, José A. García-Erce*, F. Martínez, R. Cardona, L. Pérez-Serrano, Manuel Muñoz**
Departments of Orthopaedic and Trauma Surgery and *Haematology, University Hospital “Miguel
Servet”, Zaragoza, Spain, and **GIEMSA, School of Medicine, University of Málaga, Málaga, Spain.
This study has been approved by the Institutional Review Board of the University Hospital “Miguel
Servet”, Zaragoza, Spain, and all patients gave informed consent to enter the study.
CONFLICT OF INTEREST STATEMENT
We declare that no benefits or founds were received in support of the study
* Ethical Statement
Nadia Rosencher, MD
27 rue du Fbg St. Jacques
75014 Paris (France)
Elvira Bisbe, MD
Sant Josep de la Muntanya, 12
C. M. Andrews, FRCS Orth
North Yorkshire YO12 6QL
* List of Potential Reviewers
Manuscript for THE INTERNATIONAL JOURNAL OF SURGERY
PREOPERATIVE HAEMATINICS AND TRANSFUSION PROTOCOL REDUCE THE NEED FOR
TRANSFUSION AFTER TOTAL KNEE REPLACEMENT
Jorge Cuenca, José A. García-Erce*, Fernando Martínez, Rafael Cardona, Luís Pérez-Serrano,
Department of Orthopaedic and Trauma Surgery, University Hospital “Miguel Servet”, Avenida Isabel
la Católica, 1-4, 50008-Zaragoza, Spain.
*Department of Haematology, University Hospital “Miguel Servet”, Isabel la Católica, 1-4, 50008-
**GIEMSA, School of Medicine, University of Málaga, Boulevard Louis Pasteur s/n, 29071-Málaga,
Running title: Reducing blood transfusion after knee replacement
Prof. M. Muñoz
Facultad de Medicina
Universidad de Málaga
Boulevard Lois Pasteur, s/n
Phone: +34 952 131540
Fax: +34 952 131534
Background. Unilateral total knee replacement (TKR) can result in a substantial blood loss and 30-
50% of these patients receive allogeneic blood transfusion (ABT), this transfusion rate may be even
higher among anaemic patients.
Patients and Methods. We assessed the requirements for ABT in 156 consecutive patients
undergoing surgery for primary TKR, who received iron ferrous sulphate (256 mg/day; 80 mg of Fe2+),
vitamin C (1000 mg/day) and folic acid (5 mg/day) during the 30-45 days preceding surgery, and who
were transfused if Hb <80 g/L and/or clinical signs/symptoms of acute anaemia/hypoxemia (Group 2).
A previous series of 156 TKR patients serves as a control group (Group 1).
Results. Compared to those in the control group, patients in the study group presented a lower
transfusion rate (5.8% vs. 32%, for study and control group, respectively; p<0.01), and a lower
transfusion index (1.78 ± 0.44 vs. 2.22 ± 0.65 units per transfused patient, respectively; p<0.05). After
patient’s stratification according to a preoperative Hb above or below 130 g/L, the differences in
transfusion rate remained significant, although 19% of patients from group 2 still needed ABT if their
preoperative Hb <130 g/L
Conclusion. This protocol seems to be effective for avoiding ABT in non-anaemic TKR patients,
whereas for anaemic patients another blood saving strategy, such us preoperative erythropoietin
administration or postoperative blood salvage, should be added to further increase its effectiveness.
Key words: knee replacement; allogeneic transfusion; transfusion protocol; anaemia; oral iron.
Unilateral total knee replacement (TKR) can result in a substantial blood loss (1,2) and 20-50% of
these patients receive allogeneic blood transfusion (ABT) (2-4). However, allogeneic blood is a scarce
and increasingly expensive resource and ABT is not a risk free therapy for orthopaedic patients (2,3).
All these have prompted the development of different methods to reduce or avoid ABT in these
patients, including implementation of restrictive transfusion protocols (5), use of postoperative
autotransfusion (1,6,7) and administration pharmacological agents (8).
In the one hand, a recent meta-analysis of randomised clinical trials concluded that preoperative
autologous blood donation (PABD) (RR: 0.16; 95%CI: 0.07 – 0.36) and perioperative cell salvage (RR:
0.35; 95%CI: 0.21 – 0.52) decrease the frequency of exposures to ABT when compared with a control
in orthopaedic surgery (9). However, the real contribution of autologous blood to ABT reduction
decreases if a transfusion protocol is adopted (9). From these data it can be inferred that a transfusion
protocol does in itself reduce by 25-30% the relative risk for ABT and, therefore, it must be the first
strategy to include in a blood saving program. In the other hand, it is well known that preoperative Hb
level is one of the strongest predictors for postoperative ABT after TKR (4,10,11), and that stimulation
of erythrocyte production may reduce the requirements for ABT in patients with mild anaemia (8).
In this work, we therefore evaluated the effectiveness of the implementation of a restrictive transfusion
protocol together with the preoperative administration of haematinics to reduce the requirements for
ABT in patients undergoing surgery for TKR, with special interest on those patients with mild anaemia
(Hb <130 g/L).
PATIENTS AND METHODS.
Patients. After approval by the Institutional Review Board, during a twelve month period, all patients
scheduled for elective primary TKR in a single institution were interviewed by the surgeon at least one
month before surgery to enter in a blood saving protocol. Patients with haematological diseases or
coagulation disorders, those under anticoagulant therapy or with known infection or malignancy at
admission and those with revision surgery or preoperative autologous blood donation were excluded.
Finally, 156 TKR patients entered the study (Group 2). A previous series of another 156 TKR patients
who met these inclusion criteria and underwent surgery before the implementation of the blood saving
protocol, served as control group (Group 1).
Data collection. A set of demographical and clinical data was gathered for all patients, including
gender, age, preoperative Hb concentration at the time of surgeon’s interview, 24 h postoperative Hb
concentration, transfusion rate (percentage of transfused patients), transfused units, transfusion index
(blood units per transfused patient), and length of hospital stay.
Surgical procedure. All patients were operated on by the same surgical team, under standardized
anaesthesia, antibiotic and antithrombotic prophylaxis, and postoperative analgesia. The same implant
(Nex-Gen®, Zimmer, USA) was used in all knees, with all components being cemented. All
procedures were performed using a pneumatic tourniquet that was deflated after wound closure, and
three closed suction drains (2 inside the joint and 1 subcutaneous), which were removed at the
second postoperative day. All patients stayed at the post-anaesthesia recovery unit for at least 2 hours
before being transferred to the ward.
Blood saving protocol. In group 2, patients received iron ferrous sulphate (256 mg/day; 80 mg of
Fe2+), vitamin C (1000 mg/day) and folic acid (5 mg/day) during the 30-45 days preceding surgery, and
referred adverse effects of the treatment were recorded. Normovolaemic patients were transfused if
their Hb felt below 80 g/L and presented clinical signs/symptoms of anaemia/hypoxemia, (e.g.,
hypotension, tachycardia, tachypnea, dizziness, fatigue, etc) or at a higher Hb if they presented
clinical signs and were at risk (e.g., coronary or valve heart disease or obstructive pulmonary disease).
In the control group, patients did not received preoperative haematinics and transfusion decisions
relied only in a Hb level below 90 g/L. All patients received oxygen therapy (2 L/min) during the first 48
postoperative hours. No other blood saving method was used in any patient.
Statistical analysis. Data were expressed as percentage (%) or as the mean ± SD (n). Pearson’s
Chi-square test or Fisher’s exact test was used for comparison of qualitative variables, and Student’s t
test or Wilcoxon’s rank test for comparison of quantitative variables. All statistics were performed with
SPSS 11.0 (Licensed to the University of Málaga, Spain) and a P value <0.05 was considered
There were not statistically significant differences between groups regarding patient’s age, gender
distribution, anaesthetic risk (ASA I-III), Hb at preoperative assessment, or length of hospital stay
(Table 1). At preoperative assessment, 20% of patients in the treatment group (31/156; 25 women and
6 men), and 17% of patients in the control group (26/156; 24 women and 2 men) presented a Hb
levels lower than 130 g/L (p=NS).
When compared to the control group, this blood saving protocol resulted in a lower percentage of
transfused patients (5.8% vs. 32%, for BSP and control group, respectively; p<0.01) and no serious
adverse effects of haematinic administration was reported by the patients. In control group, 15 patients
received transfusion on the second postoperative day or later, whereas in the treatment group all
transfusions were given with the first 24 postoperative hours. The differences in transfusion rate
remained significant after patient’s stratification according to preoperative Hb: 19.3 vs. 61.5% for Hb
<130 g/L (p < 0.01), and 2.4 vs. 26.1% for Hb >130 g/L (p<0.001), for treatment and control groups,
respectively (Figure 1). Similarly, a lower transfusion index was recorded for the treatment group when
compared to the control group (Table 1). After patient’s stratification according to preoperative Hb,
these differences remained significant only for patients with Hb <130 g/L (p < 0.05) (Figure 1). Finally,
as shown in table 1, 24h postoperative Hb levels were significantly higher in the treatment group with
respect to the control group (p<0.05), and this difference was even higher after subtracting the effect
of transfusions given within the first 24 postoperative hours (100 ± 14 vs. 107 ± 16 g/L, for group 1 and
2, respectively; p<0.01).
Allogeneic blood transfusions are often necessary during and after total hip and total knee arthroplasty
because of perioperative blood loss-induced anaemia. However, the rate of postoperative infection is
significantly higher in receiving ABT than in those receiving autologous blood or in those non
transfused, most probably due to a transfusion-related immuno-depression (2,3). In addition, there are
ABT-related risks, such transfusion reactions through storage-induced mechanisms, errors in blood
administration and viral or bacterial contamination, that have led to the development of different
strategies to reduce or avoid the need for ABT.
The fundamental appreciation that transfusion threshold is one of the most significant determinant of
transfusion seems to have been lost in the clinical setting, and as a result we are probably overusing
blood transfusing after elective joint replacement (5). In this respect, although a randomized trial in
critically ill patients found that a restrictive transfusion threshold (Hb <70 g/L) was as safe as al liberal
transfusion threshold (Hb <100 g/L) (12). However, a higher transfusion threshold seems more
appropriate for surgical patients with no risk factors for ischaemia, as they have a much lower degree
of monitoring. In addition, attention needs to be pay to signs and symptoms of anemia, as they are
variable depending on the patient’s age, body temperature, medications, rate of volume loss and co-
As of January 2000, a conservative transfusion protocol (Hb threshold 90 g/L) was introduced in our
institution. Patients in the control group were managed with this conservative transfusion protocol
resulting in a 32% of patients being transfused and in a transfusion index of 2.22 units per transfused
patient. To reduce this figures, we implemented a blood saving procotol for TKR in which in a more
restrictive transfusion protocol (Hb <80 g/L) was the cornerstone. Additionally, all patients received
oral haematinics for 30-45 days prior surgery to improve erythropoiesis, as low iron stores and folate
deficiency are not uncommon among elderly patients (13,14). This protocol has proved to be useful
since the transfusion rate (5.8%) and the transfusion index (1.78 units per transfused patient) in
patients from group 2 was reduced with respect to both the previous series in our institution (Group 1)
and other published series (2-6). Moreover, for patients with preoperative Hb <130 g/L, our protocol
seems to be as effective as other more complex and expensive protocols, involving the use of
rHuEPO alone or in combination with other blood conservation methods, or a flow chart on the use of
blood transfusion, with a lower transfusion trigger (Hb < 70 g/L) (15-18).
This greater effectiveness is probably due to the stimulatory effect of haematinics on erythropoiesis
(19), although part of the observed effect could reflect changes in the general attitude towards
allogeneic transfusion. In this regard, a randomised trial on iron pre-load for major joint replacement
showed that at least 18% of patients attending for hip or knee replacement were anaemic and benefit
significantly from preoperative iron supplements over 4 weeks (13). In addition, iron supplementation
in patients without obvious anaemia protects against a fall in Hb during the immediate post-operative
period, suggesting a widespread underlying depletion of iron stores in this group despite a normal Hb
(13). However, when administered after surgery, oral iron was not effective in rising Hb levels (20-21)
since post-operative erythropoiesis is limited by the inflammatory effects of surgery on iron metabolism
(22,23). In the other hand, a population-based study (n = 1562) of older persons revealed that up to
20% of them were at high risk of folate deficiency and, consequently, they should be considered for
The stimulatory effect of preoperative haematinics may also be inferred from comparison of data
obtained both groups. As shown in Table 1, there were no differences between series in preoperative
Hb at the time of surgeon’s interview, but at 24h postoperative Hb was significantly higher in group 2.
However, transfusion index was higher in group 1 than in group 2 (Table 1), and it can be assumed
that transfusion of one packet red cell unit increases patient’s Hb by 10 g/L and that perioperative
blood loss was similar in both series. Thus, by subtracting the effect of transfusion given within the first
24 postoperative hours, the preoperative treatment would account for a reduction in postoperative Hb
drop of about 7 g/L (100 ± 14 vs. 107 ± 16 g/L, for group 1 and 2, respectively; p<0.01).This effect
seems to be important as 15 patients from group 1 received transfusion on the second postoperative
day or later, whereas all patients in group 1 where transfused within the first 24h postoperative hours.
In addition, the differences in transfusion rate would have been remained if the same transfusion
threshold (Hb <90 g/L) would have been used in both groups (32 vs. 12%, for group 1 and 2,
respectively; p<0.01). Hence, as opposite to postoperative oral iron (20,21), preoperative haematinic
supplements including oral iron may enhance erythropoiesis (13) and, within a coordinated blood
saving strategy, it may help to reduce postoperative transfusion requirements (19).
In conclusion, we believe that the use of a restrictive transfusion trigger plus preoperative
administration of oral haematinics is effective for reducing allogeneic in non anaemic TKR patients.
However, the use of intravenous iron might also be considered for those patients who do not tolerate
oral iron and when time to surgery is too short for oral therapy (24,25). Finally, since 19% of the
anaemic TKR patients still needs ABT, it becomes evident that some additional blood saving method,
such us postoperative blood salvage (1,6) or preoperative administration of recombinant human
erythropoietin (26), should be associated to reduce further the need for ABT.
1. Kumar GS, Von Arx OA, Pozo JL. Rate of blood loss over 48 hours following total knee
replacement. Knee 2005; 12: 307-9.
2. Rosencher N, Kerkkamp HE, Macheras G, et al. Orthopedic Surgery Transfusion
Hemoglobin European Overview (OSTHEO) study: blood management in elective knee
and hip arthroplasty in Europe. Transfusion 2003; 43: 459-469.
3. Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE, Tooms RE, Welch RB. An analysis
of blood management in patients having a total hip or knee arthroplasty. J Bone Joint Surg
Am 1999; 81-A: 2-10.
4. Bong MR, Patel V, Chang E, Issack PS, Hebert R, Di Cesare PE. Risks associated with
blood transfusion after total knee arthroplasty. J Arthroplasty 2004; 19: 281-287.
5. Spencer J, Thomas SR, Yardy G, Mukundan C, Barrington R. Are we overusing blood
transfusing after elective joint replacement?. A simple method to reduce the use of a
scarce resource. Ann R Coll Surg Engl 2005; 87: 28-30.
6. Muñoz M, Ariza D, Garcerán MJ, Gómez A, Campos A. Benefits of postoperative shed
blood reinfusion in patients undergoing unilateral total knee replacement. Arch Orthop
Trauma Surg 2005; 125: 385-9.
7. Rees JE, Jeavons R, Dixon JH. An economic justification for autologous blood re-infusion
in primary total knee replacement surgery. Ann R Coll Surg Engl 2005; 87: 102-5.
8. Spahn DR, Casutt M (2000). Eliminating blood transfusions. New aspects and
perspectives. Anesthesiology 93: 242-55.
9. Carless P, Moxey A, O’Connell D, Henry D. Autologous transfusion techniques: a
systematic review of their efficacy. Transf Med 2004; 14: 123-44.
10. Salido JA, Marín LA, Gómez LA, Zorrilla P, Martínez C. Preoperative hemoglobin levels
and the need for transfusion after prosthetic hip and knee surgery: analysis of predictive
factors. J Bone Joint Surg Am 2002; 84-A: 216-20
11. García-Erce JA, Solano VM. Cuenca J, Ortega P. [Preoperative hemoglobin as the only
predictive factor of transfusional needs in knee arthroplasty]. Rev Esp Anestesiol Reanim
2002; 49: 254-60.
12. Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, et al. A
multicenter, randomized, controlled clinical trial of transfusion requirements in critical care.
Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials
Group. N Engl J Med 1999; 340: 409-17.
13. Andrews CM, Lane DW, Bradley JG. Iron pre-load for major joint replacement. Transfus
Med 1997; 7: 281-6.
14. Clarke R, Refsum H, Birks J, et al. Screening for vitamin B-12 and folate deficiency in older
persons. Am J Clin Nutr 2003; 77:1241-7
15. Couvret C, Laffon M, Baud A, Payen V, Burdin P, Fusciardi J. A restrictive use of both
autologous donation and recombinant human erythropietin is an efficient policy for primary
total hip or knee arthroplasty. Anesth Analg 2004; 99: 262-71.
16. Kourtzis N, Pafilas D, Kasimatis G. Blood saving protocol in elective total knee
arthroplasty. Am J Surg 2004; 187: 261-7.
17. Weber EWG, Slappendel R, Hémon Y, et al. Effects of epoetin alfa on blood transfusions
and postoperative recovery in orthopaedic surgery: the European Epoetin Alfa Surgery
Trial (EEST). Eur J Anaesth 2005; 22: 249-57.
18. Müller U, Exadaktylos A, Roeder C, Pisan M, Eggli S, Jüni P. Effect of a flow chart on the
use of blood transfusions in primary total hip and knee replacement: prospective before
and after study. Br Med J 2004; 328: 934-8.
19. Goodnough LT, Shander A, Spivak JL, Waters JH, Friedman AJ, Carson JL, et al.
Detection, evaluation, and management of anemia in the elective surgical patient. Anesth
Analg 2005; 101: 1858-61
20. Mundy GM, Birtwistle SJ, Power RA. The effect of iron supplementation on the level of
haemoglobin after lower limb arthropalsty. J Bone Joint Surg (Br) 2005; 87-B:213-7.
21. Sutton PM, Cresswell T, Livesey JP, Speed K, Bagga T. Treatment of anaemia after joint
replacement. J Bone Joint Surg (Br) 2004; 86-B:31-3.
22. Biesma DH, Van de Wiel A, Beguin Y, Kraaijenhagen RJ, Marx JJM. Post-operative
erythropoiesis is limited by the inflammatory effect of surgery on iron metabolism. Eur J
Clin Invest 1995; 25: 383-9.
23. Van Iperen CE, Kraaijenhagen RJ, Biesma DH, Beguin Y, Marx JJM, Van de Wiel A . Iron
metabolism and erythropoiesis after surgery. Br J Surg 1998; 85: 41-5.
24. Briges KR (2000). Prevention of iron deficiency in surgical patients. In: Transfusion
medicine and alternatives to blood transfusion, R&J Editions Medicales, Paris, 252-5.
25. Bisbe E, Rodriguez C, Ruiz A, Saez M, Castillo J, Santiveri X. [Preoperative use of
intravenous iron: a new transfusional therapy]. Rev Esp Anestesiol Reanim 2005; 52:536
26. Weber EWG, Slappendel R, Hémon Y, Mähler S, Dálen T, Rouwet E, et al. Effects of
epoetin alfa on blood transfusions and postoperative recovery in orthopaedic surgery: the
European Epoetin Alfa Surgery Trial (EEST). Eur J Anaesthesiol 2005; 22; 249-57.
Figure 1. Allogeneic transfusion in patients undergoing primary total knee arthroplasty, according
to the preoperative haemoglobin level. A. Percentage of transfused patients (*p<0.001, Group
1 vs. group 2, Fisher’s exact test). B. Transfusion index (units per transfused patient)
(***p<0.05, group 1 vs. group 2, Wilcoxon rank test).
Hb<130 g/LHb≥ ≥ ≥ ≥130 g/L
Preoperative Hb level
Transfusion index (U/pt)
Hb<130 g/L Hb≥ ≥ ≥ ≥130 g/L
Preoperative Hb level
Transfused patients (%)
Table 1. Demographic and clinical data of two series of patients undergoing surgery for total Download full-text
knee replacement with (Group 2) or without (Group 1, control) preoperative oral haematinic
supplements (Ferrous sulphate, 240 mg/day; vitamin C, 1000 mg/day; and folic acid, 5 mg/day,
for 30-45 days prior operation)and restrictive transfusion protocol. Data are expressed as mean
± SD, incidence or %.
Group 1† Group 2‡
Patients (n) 156 156
Age (year) 70 ± 5 71 ± 6
Gender (M/F) 48/108 53/103
Preoperative Hb (g/L) 139 ± 13 141 ± 13
24 h postoperative Hb (g/L) 105 ± 12 108 ± 14**
Transfused patients (n, %) 50 (32) 9 (5.8)*
Transfused units (n, %)
Transfusion index (units/ transfused patient) 2.22 ± 0.65 1.78 ± 0.44**
Length of hospital stay (days) 12 ± 4 11 ± 5
†Transfusion protocol: Hb <90 g/L. ‡Transfusion protocol: Hb <80 g/L and/or
signs/symptoms of acute anaemia/hypoxemia. * P < 0.01, treatment vs. control. ** P < 0.05,
treatment vs. control.