Plastic and Reconstructive Surgery

Anesthetic Considerations

Surgical correction of a cleft lip defect is usually performed at 2 to 3 months of age to allow sufficient time for maturation and associated abnormalities to become apparent. Preoperative assessment may reveal abnormalities such as mandibular hypoplasia in Pierre Robin sequence (PRS) ( Fig. 35.1 and E-Fig. 35.1 ) or restricted neck movement as in Klippel-Feil syndrome ( E-Fig. 35.2 ). PRS is defined as the triad of micrognathia, glossoptosis (caudally displaced insertion of the tongue), and respiratory distress in the first 24 to 48 hours after birth. The presence of other anomalies might warrant additional clinical or laboratory investigations. Cleft lip repair usually involves minimal blood loss, so for children with hematocrit values greater than 30%, additional preoperative laboratory testing is unnecessary. A sample for type and screen is usually sufficient for infants with a hematocrit value less than 30%.

The frequency of difficult airways in children with cleft lip and palate varies from 2.9% to 23%. The incidence of difficult intubation in children with bilateral cleft is greater than that with unilateral cleft. Furthermore, micrognathia is an independent predictor of a difficult airway. The incidence of difficult direct laryngoscopy is approximately 50% in children with micrognathia but only about 4% in those without. In infants and young children, micrognathia may be subtle and not always easily detected. However, the presence of microtia, which is associated with a 42% incidence of difficult intubation with bilateral compared with 2% with unilateral microtia, should prompt a closer examination of the mandible for hypoplastic growth and raise the possibility of a hemifacial microsomia or Treacher Collins syndrome. Intubation difficulty (by direct laryngoscopy) with isolated micrognathia decreases with increasing age, with the greatest difficulty presenting in infants younger than 6 months. This has been attributed to rapid growth of the mandible, which catches up to the maxilla, thereby aligning the two bones, by 2 years of age in most cases. A careful review of previous anesthetic records may forewarn of a difficult airway. In a second study that reported the frequency of difficult intubations in young infants with cleft lip, cleft palate without PRS, cleft-lip-palate and cleft palate, and cleft palate with PRS was 0, 2.7%, 10%, and 23%. Difficult airways increased with early airway and feeding problems (P < 0.0001). In the isolated cleft palate, a wider cleft was associated with a significantly more difficult laryngoscopy.

Induction of anesthesia via face mask is usually uncomplicated in infants with cleft lip and palate. Laryngoscopy should be performed using a straight blade via a right paraglossal approach (blade inserted into the pharyngeal gutter with tongue displaced to the left), taking care to avoid dropping the blade into the cleft (see also Fig. 14.28A-C ). If the mandible is hypoplastic, external laryngeal manipulation may be required to bring the larynx into view. In some centers, the tongue is sutured to either the mandible or lower lip to preclude airway obstruction in infants with PRS in the postnatal period. In such instances, the tongue cannot be displaced to the left to expose the larynx. To facilitate laryngoscopy in such cases, the tongue is first released from the lower lip using ketamine sedation followed by direct laryngoscopy. Alternatively, selection of a primary airway management strategy other than direct laryngoscopy (e.g., video laryngoscopy, fiberoptic intubation through a supraglottic airway) circumvents this problem and may be a preferable approach with a greater success rate on the first attempt (see Chapter 14 ).

A variety of tracheal tubes can be used to secure the airway for cleft lip and palate surgery, although the ideal tracheal tube is perhaps the oral Ring-Adair-Elwyn (RAE) tube, which can be fixed centrally to the chin to facilitate optimal surgical access. It should be noted that preformed tracheal tubes vary in length from bend to tip for uncuffed and cuffed tubes. Seven brands of preformed tracheal tubes were compared for the same size inner diameter tubes; the distance from the bend to the tip varied by 0 to 1 cm for oral cuffed tubes but 0 to 4 cm for uncuffed oral tubes. Of greater concern was that the variability for preformed nasal tubes was even greater; cuffed nasal tubes varied by 0 to 5.5 cm and uncuffed varied from 2 to 9 cm (bend to tip). The risk of an endobronchial intubation when a cuffed oral preformed tube replaced an uncuffed tube of the same diameter was 0%–27%, whereas when a cuffed nasal preformed tube replaced an uncuffed tube of the same size, the risk of an endobronchial intubation increased to 50%–100%. Thus the risk for endobronchial intubation varies among manufacturers and is greater with nasal than oral preformed tracheal tubes.

Throat packs usually impinge on the surgical field and are not normally required for cleft palate repair. Ventilation is usually controlled for the duration of the procedure (~1–2 hours). Inhalational or intravenous (IV) anesthetics combined with a short-acting opioid such as fentanyl (1–2 µg/kg) can be used for maintenance of anesthesia. Bilateral infraorbital nerve blocks may be used to provide postoperative analgesia for cleft lip repairs (see Fig. 42.12A-C and Chapter 42 videos). Such blocks reduce the need for opioids and antiemetics, improve the ability to feed, and increase parental satisfaction. Additional evidence suggests that the incidence of emergence agitation is reduced with the use of infraorbital nerve blocks. A combination of infraorbital and external nasal nerve blocks for pain control after cleft lip repair is an alternative.

During cleft palate surgery, the pharyngeal space is reduced dramatically; postoperative nasal trumpets may be required (often placed by the surgeon) to maintain airway patency and permit suctioning the airway without damaging the palatal repair ( Fig. 35.2 and E-Fig. 35.3A and B ). At the end of surgery, the trachea is extubated after the upper airway reflexes have returned and the child is completely awake. These children are at particular risk for acute upper airway obstruction in the immediate postextubation period as a result of upper airway narrowing, edema, blood, and residual anesthetic effects. Accordingly, it is very important to extubate the trachea only when the child is completely awake. Intraoperative dexamethasone (0.5 mg/kg) may be administered to mitigate postoperative airway edema. Late postoperative edema and severe subcutaneous emphysema are additional complications. Upper respiratory tract infections are common in this age group. If airway infections are present, they should weigh heavily in favor of delaying surgery until they are resolved. Antibiotics may reduce the incidence of postoperative respiratory complications. Adverse airway events, including postoperative airway obstruction, oxyhemoglobin desaturation, bronchospasm, laryngospasm, reintubation, and unplanned postoperative intensive care unit (ICU) admission have been identified in as many as 23% of children undergoing cleft palate repair ; the presence of a craniofacial syndrome, history of preoperative airway problems, and both surgeon and anesthesiologist inexperience were significantly associated with airway complications.

Arm restraints are used in many centers to prevent suture disruption. These children are monitored for signs of upper airway obstruction during the recovery period for approximately 48 hours. As soon as the child is awake, feeding with clear fluids is allowed. Postoperative pain is managed with a combination of opioids and acetaminophen. Rectal acetaminophen administered before surgery did not reduce postoperative opioid requirements in children undergoing cleft palate repair. In part this may be attributed to local infiltration at the surgical site and the slow and variable absorption of acetaminophen by the rectal route. However, IV acetaminophen has opioid-sparing effects in these children that may be useful in certain scenarios. Sphenopalatine and infraorbital nerve blocks (see Chapter 42 , Fig. 42.12 ) with a long-acting local anesthetic can be placed at the end of the procedure to prevent pain after cleft lip and palate repair. Palatal nerve block (nasopalatine, greater and lesser palatine) or a bilateral suprazygomatic maxillary nerve block reduce postoperative pain and favor early feeding.

Children who are scheduled for elective pharyngoplasty are usually school age, having undergone cleft palate repair at an earlier age. The primary objective of this procedure is to restore velopharyngeal competence for speech development, which can be achieved by a pharyngeal flap, sphincter pharyngoplasty, or palatal lengthening (Furlow double-opposing Z-plasty palatoplasty). The anesthetic goals and management are similar to those discussed for cleft palate repair.

Airway Management

Meticulous preoperative planning and evaluation of the airway are essential, particularly for children with known or possible OSA. Upper airway obstruction following induction of anesthesia in children with midface hypoplasia should be anticipated. This is usually readily managed with a jaw-thrust/subluxation technique or insertion of an oral airway. Occasionally, however, face mask ventilation may prove challenging owing to difficulty obtaining a good mask seal. External fixator devices on the face may also present challenges in managing the airway ( E-Fig. 35.6 ), and it is essential to craft the optimal plan and make necessary preparations should problems arise (e.g., emergent hardware removal, wire cutting). A laryngeal mask is a valuable rescue tool in these scenarios and should be readily available before anesthetic induction.

In the majority of children with craniosynostosis, the anatomy of the mandible and the temporomandibular joint is normal, as are upper airway dimensions, resulting in an easy direct laryngoscopy and tracheal intubation. Rarely, mandibular hypoplasia may complicate an otherwise straightforward laryngoscopy and tracheal intubation. Abnormal neck mobility also may pose additional challenges for laryngoscopy and tracheal intubation.

Blood Loss, Coagulopathy, and Hyponatremia

Crystalloid solutions are commonly administered for minimal to moderate surgical blood loss and fluid shifts during craniosynostosis surgeries. Although lactated Ringer's solution (or Hartmann's solution) is most commonly used in North America, some advocate using normal saline solution because it may be less likely to induce hyponatremia and an acid-base disturbance than lactated Ringer's solution. However, a recent study comparing the two solutions suggests that normal saline solution is more likely to induce (metabolic) acidosis than lactated Ringer's solution in infants undergoing craniosynostosis. Additionally, hyponatremia is usually mild, self-limited, and asymptomatic. The results of large adult cohort studies have demonstrated an association between hyperchloremic metabolic acidosis after normal saline and major complications, morbidity, mortality, and increased resource use. Isotonic balanced solutions may be the best choice for crystalloid management.

Surgery for craniosynostosis is associated with the potential for cardiac arrest as a result of sudden massive blood loss or underestimated blood loss. Although these procedures are extradural, significant bleeding from the scalp and cranium can occur, especially following inadvertent tears of dural venous sinuses. The blood loss can be so rapid during the surgery that the expression “trauma in progress” is applicable. The risks of massive blood loss and the need for invasive monitoring are primarily determined by the type of surgery. In children undergoing endoscopic strip craniectomy, weight less than 5 kg, those undergoing sagittal endoscopic craniectomy, those with syndromic craniosynostosis, and earlier date of surgery are associated with blood transfusion. Some centers advocate commencing blood transfusions at the time of skin incision (particularly in infants) to prevent hemodynamic instability and the need for a rapid transfusion, but this must be tempered by whether the surgery is open or endoscopic (see later discussion).

To manage the large volume and rapidity of the blood loss, it is essential to establish large-bore peripheral venous access. Central venous access (see also Figs. 49.3 and 49.4 ), usually via the internal jugular vein, is most commonly reserved for children in whom obtaining adequate peripheral venous access proves difficult but overall has been used in a minority of cases for rapid transfusion. Strategies that are important to preclude hyperkalemic cardiac arrest include infusing blood stored less than 7 days old (see Wake Up Safe; http://www.wakeupsafe.org/Hyperkalemia_statement.pdf?201501300915 ; accessed January 7, 2015) and (most importantly) avoiding hypovolemia. If only older blood is available, it should be warmed and administered slowly through a peripheral IV line to reduce the risk that hyperkalemia is present when the blood reaches the right atrium. Estimation of ongoing blood loss can be difficult because of the use of large volumes of irrigation fluid and difficulty quantifying blood loss onto surgical drapes and gowns. Invasive arterial blood pressure monitoring and serial blood gas sampling are indicated in this type of surgery (see Fig. 49.11A-D ). A urinary catheter should be inserted to monitor urine output.

Several blood conservation strategies have been proposed for this type of surgery, including preoperative recombinant human erythropoietin, acute normovolemic hemodilution, antifibrinolytics, and induced hypotension (see Chapter 12 for a more detailed discussion). Of primary importance is meticulous surgical technique and attention to hemostasis. Bleeding from the scalp incision may be reduced by infiltration with a dilute (1 : 400,000) epinephrine-containing solution. The use of the reverse Trendelenburg position may help to decrease venous pressure and blood loss from osteotomy sites but may increase the risk of venous air embolism (with a reported frequency of 5%–80%; see later discussion). For this reason, the horizontal position is preferred.

Blood-conserving dual therapy with recombinant human erythropoietin (to optimize preoperative hematocrit) and use of a cell saver has reduced transfusion in children undergoing craniosynostosis repair. Administration of preoperative recombinant human erythropoietin, in combination with elemental iron (4 mg/kg per day orally to a maximum of 200 mg/day for 6 weeks) increases the preoperative hematocrit value and may decrease the need for autologous blood transfusion. If iron stores are at all compromised, iron therapy combined with oral vitamin C (to increase gastrointestinal absorption) should begin 3 weeks before erythropoietin therapy. Currently this technique is rarely used, likely because of high costs as well as the black box warning applied to synthetic erythropoietin drugs related to increased likelihood of major complications in adult populations receiving these drugs.

Little evidence exists to suggest that autologous blood donation decreases perioperative morbidity in craniosynostosis surgery. While infants as young as 3 months have predonated, this technique is of questionable value and should not be pursued in this population. Instead, simpler and more cost-effective techniques should be used such as meticulous surgical attention to hemostasis and administration of antifibrinolytics.

Acute normovolemic hemodilution is a labor-intensive technique in which blood is collected from the child after induction of anesthesia but immediately before surgery and replaced with an appropriate volume of crystalloid or colloid, such as 5% albumin. This technique has been used in combination with other techniques, especially preoperative erythropoietin to reduce transfusion in craniosynostosis surgery. The maximum potential blood savings with this technique is modest at best. Despite some reports of efficacy, owing to the labor-intensive nature of the technique and modest blood savings, this technique is rarely used in craniofacial surgery.

The coagulation profile and clotting factors after fresh frozen plasma (FFP) or 5% albumin during craniofacial surgery have been compared in a nonrandomized study in infants younger than 12 months of age. The increases in activated partial thromboplastin time and decreases in the plasma concentration of factors XI and XIII and antithrombin III were less after intraoperative FFP than after 5% albumin. Fibrinogen concentrations remained stable in the FFP-treated group but decreased in the albumin-treated group. A hemostatic resuscitation strategy, similar to that used in massive hemorrhage from trauma, has been applied in pediatric craniofacial surgery. With this approach, blood loss is replaced using red blood cells and FFP in a 1 : 1 ratio. The central tenet of this approach is to prevent dilutional coagulopathy (see also Chapter 12 ). At one center where blood loss frequently exceeds a blood volume, FFP and packed red blood cells from the same blood donors were used for blood loss replacement; this technique essentially eliminated coagulopathy, resulted in fewer perioperative blood donor exposures, and is most likely to be useful where blood loss is expected to approach or exceed the circulating blood volume. Recombinant factor VIIa has been used successfully for intractable hemorrhage during cranial vault reconstruction in an infant, although this was an extreme circumstance in an isolated case.

Antifibrinolytic therapy may decrease blood loss during craniosynostosis repair in children. Tranexamic acid reduces blood loss and transfusions in pediatric craniofacial surgery ; the recommended dosing regimen is a loading dose of 10 mg/kg followed by a continuous infusion of 5 mg/kg per hour. The incidence of adverse events, including seizures and thromboembolic events, in children treated with or without antifibrinolytics during craniofacial reconstructive surgery was similar.

Another antifibrinolytic, ε-aminocaproic acid (EACA), is effective in reducing bleeding in cardiac and spinal procedures. In observational studies, EACA has been associated with reduced blood loss and transfusion in craniosynostosis surgery; these findings have yet to be confirmed in a controlled study. The dosing for infants undergoing craniofacial surgery is a loading dose of 100 mg/kg followed by 40 mg/kg per hour. A recent single-center prospective study of 120 infants undergoing craniosynostosis repair used thromboelastography and the platelet fibrinogen product to guide the administration of blood products. Using multivariate analysis and receiver operating curves to assess four parameters: K-time >2 : 1, MA <55 mm, α-angle <62 degrees, and the platelet-fibrinogen product <343, infants with all four predictors had a 92% probability of a blood loss of 60 mL/kg or more, whereas those with none of these parameters had an approximately 10% probability.

Induced hypotension, defined as a 10% to 20% reduction in the mean arterial blood pressure, may decrease intraoperative surgical blood loss and operating time, although studies demonstrating its effectiveness and safety during craniosynostosis surgery are lacking. The lower limits of safe blood pressure reduction in infants are unknown. A variety of pharmacologic agents have been used to induce hypotension, including inhalational agents, vasodilators, β-blockers, and remifentanil. Invasive arterial pressure monitoring is essential whenever hypotensive anesthesia is used. Induced hypotension should be used with great caution in the presence of increased ICP because of the risk of compromising cerebral perfusion pressure (i.e., the difference between mean arterial pressure [MAP] and either ICP or central venous pressure, whichever is greater). The risks of inadequate cerebral and end-organ oxygen delivery during hypotension are magnified by coexistent anemia. It is considered prudent to maintain normovolemia and normocapnia when induced hypotension is used (see Chapter 12 ). Many practitioners have determined that the potential benefits (reduction of blood loss, shorter surgery) of this technique are outweighed by the risks (cerebral ischemia and irreversible brain injury, blindness, end-organ damage). Rather than an approach of induced hypotension, some use a strategy of deliberate normotension, where the anesthetic is tailored to avoid blood pressures greater than baseline.

Whether used alone or in combination, the preceding techniques seldom obviate the need for any blood transfusions during craniofacial surgery. In a retrospective review of 60 children who underwent craniofacial surgery at a single institution, half of the children required fewer transfusions and had a reduced length of stay that were attributed to the preoperative use of iron and erythropoietin, the use of a blood-recycling device intraoperatively, and a reduced minimum hemoglobin concentration for transfusion (<7 mg/dL) compared with a placebo group. Given the potential for acute large-volume blood loss IV access with large-bore catheters remains essential and at least 2 units of packed red blood cells (PRBCs) should be crossmatched and available in the operating room at all times. IV fluids should be administered via a fluid warmer to prevent hypothermia. Maintaining normothermia preserves coagulation function and theoretically reduces bleeding and transfusion-related complications. Based on adult data, coagulopathy from dilution of soluble clotting factors on average develops when 142% of the circulating blood volume has been lost and replaced with PRBCs and crystalloid, and with thrombocytopenia developing after an average of 2.3 blood volumes were lost (see also Chapter 12 ). Estimating the blood loss based on the number of blood volumes of blood products administered, physical estimates of blood loss, and an assessment of hemostasis in the surgical field are useful empiric and clinical indicators for the need for hemostatic blood product administration. Serial determinations of the international normalized ratio prothrombin time (INR), partial thromboplastin time (PTT), platelet count, and fibrinogen concentration and the use of thromboelastography help to identify coagulopathy and guide hemostatic blood product administration.

Endoscopic repair of craniosynostosis has become a rapidly growing surgical approach to reduce bleeding and decrease morbidity. Independent risk factors for bleeding during endoscopic strip craniectomy include low body weight (<5 kg), sagittal suture surgery (related to proximity to the sagittal venous sinus), syndromic craniosynostosis, and earlier date of surgery.

Hyponatremia and cerebral salt-wasting syndrome are associated with craniosynostosis repair. Both intraoperative and postoperative hyponatremia have been described, with the latter occurring in approximately 30% of children. In a retrospective review of a cleft palate and craniofacial database, postoperative hyponatremia was associated with preoperative increased ICP, blood loss, and female gender with normal preoperative ICP. The average reduction in sodium concentration was more pronounced in children who received hyponatremic (hypotonic) (5% dextrose and 0.2% or 0.5% NaCl) compared with normonatremic (isotonic) postoperative IV fluids. The perioperative use of balanced salt solutions is recommended to prevent hyponatremia (see also Chapter 9 ).