SciELO - Scientific Electronic Library Online

vol.54 issue1Epidural abscess after patient-controlled epidural analgesia: case reportInfluence of intravenous clonidine in the cost of sevoflurane anesthesia for outpatient middle ear procedures author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Anestesiologia

Print version ISSN 0034-7094On-line version ISSN 1806-907X

Rev. Bras. Anestesiol. vol.54 no.1 Campinas Jan./Feb. 2004 



Acute normovolemic hemodilution in children submitted to posterior spinal fusion*


Hemodilución normovolémica aguda en niños sometidos a artrodesis de columna vertebral por la vía posterior



Gizelda S de Oliveira, M.D.I; Sérgio Bernardo Tenório, TSA, M.D.II; Débora O Cumino, TSA, M.D.III; Daniela B Garcia Gomes, TSA, M.D.III; Edson N Namba, M.D.IV; José Luis A Maidana, M.D.IV; Luiz Eduardo Munhoz da Rocha, M.D.V

IResponsável pelo Serviço de Anestesiologia do Hospital Infantil Pequeno Príncipe e Anestesiologista do Hospital de Clínicas da UFPr
IIProfessor Adjunto da Disciplina de Anestesiologia da UFPr e Anestesiologista do Hospital Infantil Pequeno Príncipe
IIIMembro do Serviço de Anestesiologia Pediátrica do Hospital Infantil Pequeno Príncipe
IVMembro do Serviço de Anestesiologia do Lucile's Children Hospital, Universidade de Stanford
VOrtopedista do Hospital Infantil Pequeno Príncipe e do Hospital de Clínicas da UFPr





BACKGROUND AND OBJECTIVES: Acute Normovolemic Hemodilution (ANH) is a simple and low-cost autologous transfusion method. This study aimed at evaluating whether ANH is able to decrease homologous transfusions in children undergoing posterior spinal fusion and at verifying hemodilution-induced complications in those patients.
METHODS: Participated in this study 25 children, physical status ASA I and II, submitted to posterior spinal fusion and receiving ANH (group H). Hematocrit values were recorded at surgery beginning and completion, at PACU and hospital discharge. The number of children receiving homologous transfusions during and after surgery, as well as intra and postoperative complications were also recorded. Group H data were compared to other group of children submitted to the same surgical technique, however without ANH (group S).
RESULTS: Median age and weight for both groups were, respectivelly 13 and 12 years and 41.5 and 34 kg. Immediately after anesthetic induction, 523 ml (mean) of blood were collected from group H, or the equivalent to 17.1% of volemia. Simultaneously, lactated Ringer's solution was started to decrease hematocrit to 28.8% ± 3.72%. At surgery completion and after autologous blood infusion, hematocrit has reached mean values of 27% in group H and 30.4% in Group S (p = 0.01). At PACU and hospital discharge, differences between hematocrits were not statistically significant. Homologous blood was transfused in 28% of Group H children and 79% of Group S children (p = 0.001). There has been arterial hypotension in 28% of Group H patients and 37.5% of Group S patients (p = 0.9). Four Group S patients had severe postoperative infectious complications. Hospital stay for Groups H and S was, respectively 7.56 ± 3.203 days for group H, and 9.75 ± 4.245 days for group S (p = 0.009). Group H has received 3.948 ± 1.334 ml lactated Ringer's and group S has received 2.234 ± 953 (p = 0.0001).
CONCLUSIONS: Normovolemic hemodilution is a safe method to decrease homologous blood needs in children submitted to posterior spinal fusion. There have been no anemia-related complications.

Key Words: ANESTHESIA, Pediatric; BLOOD: normovolemic hemodilution; SURGERY, Orthopedic


JUSTIFICATIVA Y OBJETIVOS: La hemodilución normovolémica aguda (HNA) es un método de transfusión autóloga simples y de bajo costeo. El objetivo de este estudio fue verificar se la HNA puede reducir la necesidad de sangre homólogo en niños sometidos a artrodesis de la columna vertebral por vía posterior y evaluar las complicaciones resultantes de la hemodilución en ese tipo de paciente.
MÉTODO: Participaron del estudio 25 niños, estado físico ASA 1 y 2, sometidos a artrodesis de columna por la vía posterior, y que recibieron HNA (grupo H). Fueron registrados los valores del hematócrito en el inicio y fin de la cirugía, en el alta de la terapia intensiva y del hospital, el percentual de niños que recibieron transfusión homóloga durante y después de la cirugía y las complicaciones ocurridas en el intra y pos-operatorio. Los datos del grupo H fueron comparados con otro grupo de niños sometidos al mismo tipo de cirugía, igual técnica anestésica, sin embargo sin HNA (grupo S).
RESULTADOS: Las medias de edad y peso de los dos grupos H y S fueron respectivamente 13 y 12 años y 41,5 y 34 kg. Fueron retirados 523 ml de sangre del grupo H, en el início de la cirugía, equivalente a 17,1% de la volemia, y reinfundida simultaneamente en solución de Ringer con lactato, reducindo el hematócrito para 28,8% ± 3.72%. En el final de la cirugía, después de la reinfusión de la sangre autóloga, el hematócrito chegó al valor médio de 27% en el grupo H y 30,4% en el grupo S (p = 0,01). En el alta de la unidad de terapia intensiva y en el alta del hospital, las diferencias entre los hematócritos no eran estadisticamente significativas. Transfusión de sangre homóloga fue utilizada en 28% de los niños del grupo H y 79% del grupo S (p = 0,001). Hipotensión arterial ocurrió en 28% de los pacientes del grupo H y 37,5% del grupo S (p = 0,9). En el grupo S, 4 pacientes tuvieron complicaciones infecciosas importantes en el pos-operatorio. La duración del internamiento hospitalar en los grupos H y S fueron, respectivamente, 7,56 ± 3,203 días y 9,75 ± 4,245 días (p = 0,009). El grupo H recibió 3.948 ± 1.334 ml de Ringer con lactato y el grupo control 2.234 ± 953 ml (p < 0,0001).
CONCLUSIONES: La HNA es um método seguro y capaz de reducir la necesidad por sangre homólogo en niños sometidos a artrodesis de columna. No hubo complicaciones decurrentes de la anemia.




Blood transfusion is knowingly a risky therapy 1. Major risks are infectious diseases transmission, immune defenses decrease, non-cardiogenic pulmonary edema and anaphylactic reactions 2-6, in addition to human error in the process of blood collection, typing, storage and infusion 7.

There are currently no substitutes for some of the most important blood functions, such as coagulation and oxygen transportation, so blood transfusions continue to be widely used, especially in major surgical procedures 8.

Autologous transfusion is the technique that uses blood from patients themselves. Its major advantage is to decrease risks associated to homologous transfusion. Three autologous transfusion methods are: pre-donation, collected blood reinfusion in the surgical field, and acute normovolemic hemodilution (ANH). As compared to the others, the latter has the advantages of being technically simpler, not requiring sophisticated equipment and having a lower cost 2,9. In spite of all advantages, ANH is still seldom used 10,11

ANH is the removal, soon after anesthetic induction, of a predetermined volume of blood, which is reinfused during surgery or, preferably, at its completion. Volume is maintained by the simultaneous replacement of crystalloids, colloids, or the association of both 12,13.

This study aimed at evaluating whether ANH could decrease homologous blood needs in children submitted to posterios spinal fusion and at evaluating hemodilution-related complications in those patients.



After the Hospital Ethics Committee approval and families' informed consent, participated in this study 25 children aged 7 to 17 years, physical status ASA I and II, scheduled for posterior spinal fusion and acute normovolemic hemodilution. Exclusion criteria were children below 7 years of age and hematocrit below 30%.

Excessively anxious patients were premedicated with 0.5 midazolam. Anesthesia was induced with propofol or sevoflurane and N2O and was maintained with isoflurane, N2O, fentanyl and atracurium. Patients were monitored by diuresis, electrocardiography, pulse oximetry, capnography, anesthetic gases analyzer, invasive and noninvasive blood pressure and temperature. Children were volume-controlled ventilated with tidal volume varying 7 to 10, with the Shogun Evolution® anesthesia machine. All patients were awaken during surgery after vertebral fusion to check possible spinal cord compression (wake-up test).

Autologous blood was collected from the radial artery after anesthetic induction. Blood volume was measured by weight; blood bag was maintained on a precision scale during collection; each gram was considered as representing 1 ml blood. Anticoagulant volume was a function of the theoretical volume of blood to be collected. Simultaneously, lactated Ringer's solution was infused through a peripheral vein in a volume equivalent to 3 times the volume of collected blood. After blood collection, patients were placed in the prone position with pads under shoulders and pelvis.

Autologous blood was reinfused at surgery completion, after stabilization of surgical blood losses, or during surgery, if there were major hemodynamic changes, such as arterial hypotension, tachycardia, hypoxemia or hematocrit below 20%. At surgery completion, all patients received diuretics and were referred to intensive care unit under mechanical ventilation.

Autologous blood volume obtained was calculated by the following formula 14:

Ladm: admissible blood loss;
V: volemia, calculatedd based on the formula weight x 70;
Hct i: initial hematocrit;
Hct f: final hematocrit.

For the purpose of this study, arterial hypotension was defined as more than 30% decrease in baseline pressure. Blood volume was calculated to reach 30% hematocrit. Hematocrit was evaluated in the preoperative period, at surgery completion, at intensive care unit discharge and at hospital discharge.

Data were compared to a group of children aged 7 to 16 years, physical status ASA I and II who, two years ago had been submitted to the same surgical procedure, same anesthetic approach and same transfusion criteria, only differing in hemodilution, which was not performed. The group of children submitted to hemodilution was called Group H and the other group was called Group S.

Parametric data were summarized by mean or median when dispersion measured by standard deviation was high. Proportion Analsys was used to compared binominal data. Student's t test was used to analyze quantitative data or Mann-Whitney test was used when sample standard deviation was higher than 30% of mean, considering statistically significant p < 0.05.



Group H aged 7 to 17 years (median 13) and weighed 19 to 62 kg (mean 41.5). Group S aged 7 to 16 years (median 12) and weighed 12 to 70 kg (median 34). In the beginning of surgery, 523 ml blood was collected from group H, equivalent to 17.1% volemia. With lactated Ringer's infusion, hematocrit has decreased to 28.8% ± 3.72%. Hematocrit means were only statistically significant between groups at surgery completion. Homologous blood transfusion was needed in 28% of Group H patients and 79% of Group S patients (p = 0.001). At surgery completion, Group H hematocrit has varied 18% to 33% and Group S hematocrit has varied 18% to 40%. There has been arterial hypotension in 28% of Group H and 37.5% of Group S patients (p = 0.99). In the control group, 4 patients had major postoperative infectious complications. Two of these children have remained admitted for 11 and 21 days, respectively. Hospital stay for Groups H and S were, respectively, 7.56 ± 3.203 days and 9.75 ± 4.245 days (p = 0.009). Group H has received 3.948 ± 1.334 lactated Ringer's and Group S 2.234 ± 953 ml (p < 0.0001). Demographics data, hematocrit results at surgery beginning and completion, PACU and hospital discharge are shown in table I.



Our results have shown that ANH may decrease the need for homologous blood in children undergoing posterior spinal fusion. The observed decrease in blood transfusion (28% x 79%) is clinically important because spinal surgeries often need blood transfusions. A study with 30 adolescents submitted to posterior spinal fusion has estimated surgical blood losses in 76 equivalent to more than one volemia 15. ANH effectiveness in decreasing homologous blood needs during surgeries is documented in other studies. A group of 27 adolescents submitted to spinal fusion due to idiopathic scoliosis, who receved ANH (mean 8.2% hematocrit decrease), needed less homologous transfusions during surgery as compared to the control group 16. Homologous transfusion was only used in 6 out of 25 children aged 25 days to 10 years, submitted to major oncologic surgeries and hemodiluted until mean 14.3% hematocrit 17. From an 8-month old child with estimated 800 ml volemia, 160 ml blood were collected and produced 17% hematocrit. Although being a major abdominal surgery, homologous blood was not used 18.

Hemodilution role in decreasing blood transfusion needs derives from the fact that red cell losses are decreased during surgery due to hematocrit decrease 19. So, less red cells are lost because there are less red cells in circulating blood to be lost. Although this theory is questioned by at least one mathematical model 20, it is still accepted by many investigators.

No serious complication may be charged to anemia, in spite of low hematocrit observed in most hemodiluted children and in some without hemodilution.

It is worth reminding that mean hematocrit at surgery completion in Group H was 27%, which means half the children had hematocrits below 27%. Since hemoglobin is responsible for transporting almost all oxygen consumed by tissues, it is true that compensation mechanisms have been used by the body. In fact, acute anemia is associated to increased cardiac output, increased oxygen extraction by tissues and to the right shift of hemoglobin dissociation curve 21.

Association between hemoglobin and increased cardiac output has been shown in experimental and clinical studies. Dogs submitted to hematocrit decrease from 42.2% to 10.6% had their cardiac output increased from 4 to 5.2 L.min-1.m2. However, animals presented major blood pressure decrease and myocardial ischemia when hematocrit has reached 20% and 11.4%, respectively 22. These findings were proven by human studies 23.

Blood viscosity decrease is another factor atenuating anemia in hemodiluted patients. It is known that vascular resistance in small veins is inversely proportional to flow rate when hematocrit value is close to normal. In practice, this means increased vascular resistance in capillary circulation. Blood dilution puts it close to a Newtonian fluid, which is defined as a fluid with linear ratio between rate and resistance. Hemodilution, then, might be advantageous in decreasing microcirculation vascular resistance, where blood circulation is, in general, slower 24.

Children have hemodynamic response similar to adults when hemodiluted. From a group of 6 children with mean age of 8.6 years and mean weight of 27.7 kg, all with malignant tumors and children of Jeovah's Witnesses parents, mean 618 ml autologous blood were collected at surgery beginning, equivalent to approximately 22.3% volemia. Hematocrit was decrased from 38% to 25%. Simultaneously, cardiac index has increased 24.3% and systemic vascular resistance has decreased 21.5%. Additional blood losses during surgery have caused mean hematocrit decrease to 16%, resulting in additional cardiac index increase and suggesting a cause-effect relationship between these two variables 25.

Hemodiluted children remained less time in the hospital as compared to non-hemodiluted children (7.56 x 9.75 days). There are no elements in our study to associate this difference to hemodilution. There is, however, a factor increasing mean hospitalization time in Group S, which is the presence of 4 postoperative infections. At least two children had to remain admitted for a longer time as compared to mean hospitalization time of remaining children (11 and 21 days). This has contributed for a statistically significant difference between groups. All four patients with infection had received homologous blood. As already said, it is impossible to conclude from our data that those infections were caused by the blood, however, there are several studies in the literature proving that homologous blood is immunesuppressant 3,4,26-29.

One limitation of this study and of others published in the literature was the comparison with a historical control group. Other studies, preferably prospective and randomized, about effects and repercussions of hemodilution in children could confirm our findings and bring more information about the subject.

In the conditions of our study, it has been shown that acute normovolemic hemodilution is a simple and safe autologous transfusion method able to decrease homologous blood needs in posterior spinal fusion surgeries, without anemia-related complications.



01. Dodd RY - The risk of transfusion-transmitted Infection. N Engl J Med, 1992;327:419-420.        [ Links ]

02. Spahn D, Casutt M - Eliminating blood transfusions. Anesthesiology, 2000;93:242-255.        [ Links ]

03. Triulzi DJ, Vanek K, Ryan DH et al - A clinical and immunologic study of blood transfusion and postoperative bacterial infection in spinal surgery. Transfusion, 1992;32:517-524.        [ Links ]

04. Avall A, Hyllner M, Bengtson JP et al - A postoperative inflammatory response after autologous and allogenic blood transfusion. Anesthesiology, 1997;87:511-516.        [ Links ]

05. Crosby ET - Perioperative haemotherapy: II. Risks and complications of blood transfusion. Can J Anaesth, 1992;39:822-837.        [ Links ]

06. Wolf CFW, Canale VC - Fatal pulmonary hypersensitivity reaction to HL-A incompatible blood. Transfusion: report of a case and review of the literature. Transfusion, 1976;135-140.        [ Links ]

07. Myhre BA - Fatalities from blood transfusions. JAMA, 1980;244: 1333-1335.        [ Links ]

08. Spiess BD, Counts RB, Gould AS - A History of Transfusion em: Perioperative Transfusion Medicine. Williams & Wilkins, Baltimore 1998;9-10.        [ Links ]

09. Goodnough LT, Brecher ME, Kanter MH et al - Transfusion Medicine, Part I. N Engl J Med, 1999;340:438-444.        [ Links ]

10. Bryson GL, Laupacis A, Wellls GA - Does acute normovolemic hemodilution reduce perioperative allogeneic transfusion? A meta-analysis. Anesth Analg, 1998;86:9-15.        [ Links ]

11. Loubser PG, De Juan R - Survey of acute normovolemic hemodilution utilization by Anesthesiology programs. Anesthesiology, 1998:A-1193.        [ Links ]

12. Ereth MH, Oliver Jr W, Santrach PJ - Intraoperative Techniques to Conserve Autologous Blood: Red-Cell Salvage, Platelet-Rich Plasma, and Acute Normovolemic Hemodilution, em: Spiess BD, Counts RB, Gould Steven A et al - Perioperative Transfusion Medicine, Baltimore, Williams & Wilkins, 1998;336-342.        [ Links ]

13. Gillon J, Thomas MJG, Desmond MJ - Acute normovolaemic haemodilution. Transfusion, 1996;36:640-643.        [ Links ]

14. Stehling L, Zauder HL - Controversies in transfusion Medicine - Perioperative hemodilution: Pro. Transfusion, 1998;34: 265-268.        [ Links ]

15. Guay J, Haig M, Lortie L et al - Predicting blood loss in surgery for idiopathic scoliosis. Can J Anaesth, 1994;41:775-781.        [ Links ]

16. Du Toit, Relton JES, Gillespie R - Acute hemodilutional autotransfusion in the surgical management of scoliosis. J Bone Joint Surg, 1978;60:178-180.        [ Links ]

17. Schaller RT, Schaller J, Morgan A et al - Hemodilution anesthesia: a valuable aid to major cancer surgery in children. Am J Surg, 1983;146:79-84.        [ Links ]

18. Kraft M, Dedrick D, Goudsouzian N - Hemodilution in an eight-month infant. Anaesthesia, 1981;36:402-404.        [ Links ]

19. Goodnough LT, Brecher ME, Kanter MH et al - Transfusion Medicine, Part II. N Engl J Med, 1999;340:525-533.        [ Links ]

20. Feldman JM, Roth JV, Bjoraker DG - Maximum blood sabing by acute normovolemic hemodilution. Anesth Analg, 1995;80: 108-113.        [ Links ]

21. Robertie PG, Gravlee TP - Safe limits of isovolemic hemodilution and recommendations for erythrocyte transfusion. Intern Anesth Clin, 1990;28:197-204.        [ Links ]

22. Tambara EM - Avaliação da Hemodiluição Normovolêmica - Estudo Experimental em Cães. Tese Apresentada no Curso de Pós-Graduação em Clínica Cirúrgica do Setor de Ciências da Saúde da Universidade Federal do Paraná, 1996.        [ Links ]

23. Laks H, Pilon RN, Klovekorn Anderson W et al - Acute hemodilution: its effect on hemodynamics and oxygen transport in anesthetized man. Ann Surg, 1974;180:103-109.        [ Links ]

24. Bruder N, Cohen B, Pellissier D et al - The effect of hemodilution on cerebral blood flow velocity in anesthetized patients. Anesth Analg, 1998;86:320-324.        [ Links ]

25. Iterson MV, Waart FJMV, Erdmann W et al - Systemic haemodynamics and oxygenation during haemodilution in children. Lancet, 1995;346:1127-1129.        [ Links ]

26. Ford CD, Van Moorleghem G, Menlove R - Blood transfusions and postoperative wound infection. Surgery, 1993;113: 603-607.        [ Links ]

27. Agarwal N, Murph JG, Cayten CG - Blood transfusion increases the risk of infection after trauma. Arch Surg, 1993;128:171-176.        [ Links ]

28. Mezrow CK, Bergstein I, Tartter PI - Postoperative infections following autologous and homologous blood transfusions. Transfusion, 1992;32:27-30.        [ Links ]

29. Landers DF, Hill GE, Wong KC et al - Blood transfusion-induced immunomodulation. Anesth Analg, 1996;82:187-204.        [ Links ]



Correspondence to
Dr. Sérgio Bernardo Tenório
Address: Rua Dr. Aluízio França, 141
ZIP: 80710-410 City: Curitiba, Brazil

Submitted for publication February 17, 2003
Accepted for publication May 16, 2003



* Received from Departamento de Cirurgia da Universidade Federal do Paraná. Trabalho realizado no Hospital Infantil Pequeno Príncipe, Curitiba, PR

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License