Preoperative

Even though the effectiveness of routine anemia screening and the implementation of specific preventive measures for children without anemia-related signs or symptoms has not been demonstrated,³⁴ high-risk neurosurgical children would benefit from the preoperative detection and treatment of anemia. This would allow for a preoperative optimization of the hemoglobin level, through the administration of oral or intravenous iron, sometimes in combination with recombinant human erythropoietin. Anemia is defined as a hemoglobin concentration less that the 5th percentile for the patient’s age.³⁵ At least 25% of preschool-age children in industrialized countries are anemic,³⁶ with a global prevalence in children being 42.6%. The incidence of anemia in hospitalized children has not been reported, but up to 37% of neonates in US hospitals are anemic.^{37, 38} Severe anemia has a prevalence of 0.9 to 1.5% and is associated with substantially worse mortality and cognitive and functional outcomes. The common etiologies of preoperative anemia in children are nutritional, iron deficiency anemia, and iatrogenic blood loss (due to frequent in-hospital blood draws). Preoperative anemia is associated with increased mortality,³⁹ and has been shown to be an independent risk factor for postoperative mortality in neonates⁴⁰ (Figure 5.2) and children.⁴¹

Figure 5.2. Distribution of preoperative hematocrit for neonates undergoing noncardiac surgery in US hospitals for survivors and nonsurvivors.

Early detection and treatment of preoperative anemia include enteral or intravenous iron therapy. The dose of iron should be calculated based on the iron daily intake, and varies between 3 to 6 mg/kg/day divided in 3 doses. The maximal dose should not exceed 150 to 300 mg per day. Erythropoiesis stimulating agents (ESAs), such as recombinant human erythropoietin, increase the hemoglobin preoperatively, and may be a useful adjuvant in certain situations such as children at higher risk of bleeding and blood product transfusion requirement (e.g., weight < 10 kg, or a syndromic patient with comorbidities). ESA clinical protocols of epoetin alfa 600 IU kg/week subcutaneously for 3–4 weeks in divided doses, with iron, vitamin B12, vitamin E, and folic acid oral supplementation, have been described.⁴² However, the risk-benefit ratio (including risks of blood transfusion vs. risks of ESA) needs to be taken into account, since no large randomized studies have assessed the safety and efficacy of ESAs in the pediatric neurosurgical population. ESAs may cause rare complications such as red blood cell aplasia, infections, seizures, possible malignancy, hypertension, and venous thrombosis, which have led to its restricted use and a Food and Drug Administration (FDA)-mandated black box warning. Transfusion of blood may be a therapeutic modality to consider, but may also be associated with some important risks. Therefore, alternative strategies should be first and foremost utilized. The frequency and volume of blood draws and invasive procedures should be limited. Reducing blood draws directly correlates with reduced transfusion.

Careful preoperative preparation and optimization such as preoperative tumor vessel or arteriovenous malformation embolization or neoadjuvant chemotherapy, when appropriate, are important preemptive strategies.⁴³ Choosing an endoscopic approach, when feasible, is less invasive than an open operation. Surgical approach planning can be supplemented with preoperative imaging and 3D models of vessels and structures to plan appropriate, less invasive surgical options. A surgical navigation system can also be used to interface with magnetic resonance imaging (MRI) and computed tomography (CT) scans for preoperative planning.

Preoperative autologous blood donation may be offered in an older child. Most programs have a minimum weight of 20 kg, with a usual donation volume between 6 and 10 mL/kg. Disadvantages with this process include issues with rendering the patient anemic preoperatively, incomplete donations, wastage, clerical error, and blood storage issues. This technique is associated with substantial risks and costs, and therefore its routine use is not recommended, and should be limited to very specific conditions (e.g., alloantibodies).⁴⁴

Intraoperative

During the intraoperative period, a number of options should be used, alone or in combination, to decrease the bleeding risk and the need for blood products transfusion.

Surgical techniques that can help control bleeding include the use of an intraoperative navigation system to localize and define vessels and lesions, minimally invasive surgery or endoscopic surgery, and the use of topical hemostatic agents. Surgical skills should be implemented to maintain meticulous hemostasis and control major vessels. Techniques to decrease bleeding in the surgical field include electric cautery, clips, careful dissection, and proper patient positioning to avoid venous congestion. Infiltration of a local vasoconstrictor and utilizing topical hemostatic agents, such as Gelfoam®, Floseal®, Fibrin Glue®, and Surgicel®, are other adjuncts to minimize local bleeding; however randomized trials to prove efficacy in this area are lacking.⁴⁵

Intraoperative blood salvage is another option that involves collecting autologous blood from the surgical site to be processed and given back to the patient during the surgery. Recently developed devices have a capacity as low as 30 mL, and are effective in reducing allogeneic blood transfusion in patients weighing as low as 10 kg.⁴⁶ The use of cell salvage is controversial in cancer surgery due to the potential risk of dissemination of tumor cells through the reinfusion of cell salvaged blood. However, recent studies have suggested that cell salvage can be used safely in cancer surgery⁴⁷ by utilizing white blood cell depletion filters and irradiating (50 Gy) cell salvage concentrate to minimize risks of tumor cell transfer. Contraindications include the use of agents that would result in red blood cell lysis (sterile water, alcohol, hydrogen peroxide), clotting agents (Surgicel and Gelfoam), contaminants (urine, bone, infection), and irrigating solutions (Betadine).⁴⁸

Acute normovolemic hemodilution (ANH) and hypervolemic hemodilution (HH) are additional strategies. ANH involves removal of the patient’s blood prior to surgery after induction of anesthesia and replacement with crystalloid or colloid to maintain normovolemia. It may be useful in older children and those who are difficult to cross match due to multiple antibodies. In younger patients this may not be feasible due to hemodynamic instability and anemia, and HH may be an option. HH involves initially hemodiluting the patient’s blood to an acceptable Hct (usually about 25%) with approximately 15 mL/kg of colloid without blood removal and then subsequent fluid restriction during the case. HH is a technique that is useful when the starting Hct is >40% (as with ESA treatment preoperatively).

Anesthetized children are particularly prone to becoming hypothermic. Hypothermia, in combination with acidosis, inevitably leads to an impaired coagulation process and may worsen bleeding. Thus, forced-air warming should be rigorously used and temperature measured continuously.

Monitoring of Coagulopathy and Point of Care

Standard coagulation has been considered as the “gold standard” for the diagnosis of congenital and acquired coagulopathies, and to guide the administration of anticoagulation both in adults and children. Activated partial thromboplastin time (PTT), international normalized ratio (INR), and prothrombin time (PT) remain widely used to assess coagulation status pre-, intra-, and postoperatively. In addition, the standard Clauss method is also used to measure the plasmatic fibrinogen concentration. Unfortunately, those tests were not designed to monitor perioperative coagulopathy or to guide the administration of hemostatic agents in bleeding situations. As it takes 30–45 minutes to obtain the results from standard coagulation assays, limited information is provided by these tests in the context of acute bleeding.⁴⁹ In addition, standard laboratory tests are performed on platelet poor plasma (PPP) and do not allow for a global assessment of coagulation, giving no information about clot firmness and clot lysis.

Over the past decade thromboelastography (TEG®) and thromboelastometry (ROTEM®) have become more popular for the assessment of coagulation and to guide the administration of blood products in bleeding patients. With the thromboelastometry method, different assays are available and allow for assessment of the intrinsic pathway (INTEM), the extrinsic pathway (EXTEM), and the fibrinogen function (FIBTEM) (Figure 5.3). Thromboelastography also allows for the assessment of the intrinsic activation and the fibrinogen function. More recently, other tests have been developed to assess platelet function. Among the large number of parameters, viscoelastic assays measure clot initiation (e.g., clotting time), clot firmness (e.g., clot amplitude measured after 5, 10, 20, 30 minutes as well as the maximal amplitude), and clot stability (e.g., percentage of lysis after 30 or 60 minutes). This approach allows for a rapid assessment of the coagulopathy, and a recent study suggested that early values of clot amplitudes measured as soon as 5, 10, or 15 minutes after clotting time could be used to predict maximum clot firmness in all ROTEM® tests.⁵⁰

Figure 5.3. Thromboelastometry (ROTEM) tracings. EXTEM (screening test for extrinsic hemostatic system) and FIBTEM (fibrinogen assay) are shown.

Viscoelastic assays are now integrated in transfusion algorithms, and their use is recommended in all European and American guidelines. In children undergoing neurosurgery, both thromboelastography⁵¹ and thromboelastometry⁵² have been used successfully. A significant decrease in plasma and platelet transfusions after the implementation of a transfusion algorithm based on thromboelastometry and the use of factor concentrates (e.g., fibrinogen concentrate and concentrate of FXIII) have recently been reported.⁵³

Antifibrinolytics

Antifibrinolytic drugs include the lysine analogs that competitively inhibit the activation of plasminogen to plasmin, tranexamic acid (TXA) and epsilon aminocaproic acid (EACA). TXA is the most common antifibrinolytic agent used since EACA is not available in many countries (such as Canada, New Zealand, and most of Europe). There is strong evidence that the implementation of a PBM strategy involving intraoperative prophylactic administration of TXA (and weaker evidence for EACA) is useful for decreasing blood loss and transfusion requirement in children undergoing major surgery¹² including craniofacial surgery.^{54, 55} TXA has been used successfully in high-risk pediatric neurosurgical tumor surgery.⁵⁶ Even though different dose schemes have been described, recent pharmacokinetic data⁵⁷ and pharmacodynamic data⁵⁸ support the use of lower doses. Based on recent data, a TXA dose of 10 mg/kg LD followed by a continuous infusion of 5 mg/kg/hr is recommended⁵⁷ (Figure 5.4). Regarding EACA, a loading dose of 100 mg/kg followed by a continuous infusion of 40 mg/kg/hr is recommended in infants undergoing craniofacial surgery.⁵⁹ A retrospective study supports the efficacy of EACA craniofacial surgery, but a well-designed prospective trial is yet to be done.⁶⁰

Figure 5.4. Tranexamic acid plasma concentration curve over time in pediatric craniofacial surgery. Dosage recommendations shown (A–D) for target tranexamic acid plasma levels 20 μg/mL and 50 μg/mL. Recommended dose of 10 mg/kg loading dose and 5 mg/kg/hr infusion for the duration of the surgery (5 hours on average).

TXA and EACA have side effects such as seizures; the incidence is low and likely dose related. Antifibrinolytics should be considered if massive blood loss is anticipated such as in hemispherectomy surgery. TXA has been used successfully in choroid plexus papilloma resection surgery, a highly vascular tumor of infancy, as part of a multimodal blood conservation strategy.⁵⁶ There may be a role for TXA in treating subarachnoid hemorrhage in the prevention of rebleeding and improvement in neurological outcome, and an ongoing trial may answer this question.^{61, 62} Investigating further the efficacy, safety, appropriate dosing scheme, and pharmacokinetic profile of antifibrinolytics in children having neurosurgical procedures with a high potential for bleeding will have a positive impact on patient care.

Fluid Management

The goal of intraoperative fluid management should be to maintain normovolemia while avoiding hypervolemia, to minimize swelling, edema, and hemodilution. Recent studies have suggested that an intraoperative positive fluid balance may increase the need for transfusion and the incidence of postoperative fluid overload, both shown to significantly affect patients’ outcomes.^63–65 Even though most of these studies were performed in children undergoing cardiac surgery, fluid overload should be avoided in children undergoing neurosurgery, and a careful monitoring of fluid administration should be recommended. Normal saline is commonly used as maintenance in intracranial neurosurgical procedures as it is mildly hyperosmolar (308 mOsm) and therefore may minimize brain edema.⁶⁶ However, if given in large volumes (> 60 mL/kg), it may lead to hyperchloremic metabolic acidosis.⁶⁷ Ringers lactate (273 mOsm) and PlasmaLyte (294 mOsm) are preferred especially when used in extracranial cases as they are associated with less severe acidosis. While some data suggested that colloid administration is associated with better hemodynamic effect with a less positive balance,⁶⁸ most of the studies were performed in children undergoing cardiac surgery and critically ill patients, and no randomized controlled trial has reported the superiority of colloids over crystalloids for perioperative fluid management in children undergoing neurosurgical procedures.²⁹ Certainly in the scenario of massive blood loss, colloids may have the advantage over crystalloids to better replace the volume lost and better restore normovolemia, preventing and treating hypotension together with other adjuncts such as vasopressors and blood product transfusion. Notably, in patients with a disrupted blood-brain barrier, caution regarding the liberal use of colloids should be mentioned, as the SAFE (Saline versus Albumin Fluid Evaluation) trial showed that in adult intensive care unit (ICU) patients, colloid administration was associated with a higher mortality in the head trauma subgroup.⁶⁹

Transfusion of Red Blood Cells

While the international blood transfusion guidelines recommend against a single transfusion trigger, maintenance of a hemoglobin (Hb) level >8 g/dL is an accepted standard intraoperatively in an actively bleeding child,^{29, 30} while in stable critically ill or anesthetized children, a threshold of 7 g/dL is sufficient.⁷⁰ This trigger is supported by recent trials reporting no difference in morbidity and mortality when a restrictive transfusion strategy was compared with a more liberal one in critically ill children.^{71, 72} Caution should be mentioned regarding neonates as the optimum transfusion Hb trigger has not been determined by evidence-based prospective trials and a higher target may be necessary.⁷³

Preoperative calculation of the circulating blood volume could help clinicians to estimate the importance of the blood loss, and help guide the administration of fluids and blood products. The blood volume of a preterm infant is 90–100 mL/kg, while the blood volume of a term neonate is 80–90 mL/kg. Above 3 months and until 1 year of age the blood volume will vary between 70 and 80 mL/kg, and will be considered as being 70 mL/kg in children older than 1 year. In addition to the estimation of the child’s blood volume, the maximum allowable blood loss (MABL) is usually calculated using the circulating blood volume (CBV), the initial hematocrit or hemoglobin, and the minimum acceptable value: MABL = CBV × (initial hematocrit – minimum acceptable hematocrit) / initial hematocrit.

This approach allows for the determination of the minimum hematocrit or hemoglobin tolerable for each patient and should be based on the child’s comorbidities, ability to maintain adequate cardiac output and oxygen delivery, metabolic needs, and tolerance to anemia. In addition to the MABL, the CBV can also be used to estimate the volume of allogeneic red blood cell needed to reach a certain hematocrit or hemoglobin level.

RBC needed = (targeted hematocrit – current hematocrit) × CBV / hematocrit of the packed red blood cell.

Transfusion volume should be limited to the volume required to achieve a certain target and to correct intraoperative acute anemia. Overtransfusion should be avoided as adverse events are directly related to the volume administered, with higher incidences of complications reported following the administration of 40 mL/kg or more.⁷⁴ As a rule of thumb, the required transfusion volume in children could also be calculated as follows: bodyweight (kg) × desired increment in Hb (g/dL) × 5. The goal is to restore/maintain tissue oxygenation to vital organs and tissues and to consider the clinical status of the patient.

Transfusion of Platelets

In children undergoing neurosurgery, a preoperative platelet count varying from >50,000 to 100,000 cells/µL is suggested based on the bleeding risk, the type of procedure, and the patient’s comorbidities. In case of active bleeding, platelet transfusion is recommended to maintain a minimal platelet count above 80,000 cells/µL. Platelet count can be expected to rise by approximately 50,000 cells/µL after transfusion of 5–10 mL/kg of an apheresis platelet concentrate. The transfusion of platelet concentrate carries the highest risk of side effects of all allogeneic blood products (bacterial contamination of platelet components is the second most common cause of transfusion-related deaths) and therefore should be performed cautiously.

Transfusion of Fresh Frozen Plasma

Accepted practice guidelines suggest replacing coagulation factor deficiency with fresh frozen plasma (FFP) when there is ongoing clinical microvascular bleeding and the PT >1.5 times normal, or INR>2, or PTT >2 times normal or correction of excessive microvascular bleeding secondary to coagulation factor deficiency in patients transfused more than one blood volume (when laboratory values can’t be obtained in a timely fashion) at a dose of 10–15 mL/kg.³⁰ However, this value may be too low, and therefore expert consensus recommends as much as 15–30 mL/kg FFP to replace coagulation factors in an actively bleeding child. Recent European guidelines highlight that while FFP transfusion is recommended to treat severe bleeding by several guidelines, high-quality evidence is lacking regarding the indication and dosing.²⁹ Certainly, severe risks have been reported including the noninfectious serious hazards of transfusion: TRALI, TACO, TRIM, and multiple organ failure.

The 2010 American Association of Blood Banks (AABB) Evidence-Based Practice Guidelines for plasma transfusion state, “We cannot recommend for or against transfusion of plasma for patients undergoing surgery in the absence of massive transfusion (quality of evidence = very low).” The AABB recommends that plasma be transfused to trauma patients requiring massive transfusion (quality of evidence = moderate). But the organization “cannot recommend for or against transfusion of plasma at a plasma: RBC ratio of 1:3 or more in trauma patients during massive transfusion (quality of evidence = low).” Furthermore these guidelines state, “It is unclear how this strategy pertains to managing massive blood loss in the pediatric surgical patient.”⁷⁵

However, prophylactic transfusion of plasma should not be recommended in a nonbleeding child and should certainly not be considered before an invasive procedure (e.g., lines removal) even in the presence of prolonged coagulation tests. Indeed, plasma remains frequently transfused in critically ill children or before a major surgery in the absence of bleeding. A recent audit of plasma transfusions in critically ill patients concluded that one-third of transfused patients were not bleeding and that plasma transfusion significantly improved the INR only in those patients with a severe derangement, indicated by an INR > 2.5.⁷⁶ Furthermore, coagulation tests are not sensitive to the increase in coagulation factors resulting from plasma transfusion. FFP is not recommended for volume expansion.