Tonsillectomy/Sleep Apnea/Hemorrhage Case File

Tonsillectomy/Sleep Apnea/Hemorrhage Case File

Lydia Conlay, MD, PhD, MBA, Julia Pollock, MD, Mary Ann Vann, MD, Sheela Pai, MD, Eugene C. Toy, MD

Case 39

A 2-year-old 14-kg girl child presents for preoperative anesthetic evaluation in the induction area just prior to surgery for tonsillectomy. Her mother reports that she has a history of obstructive sleep apnea (OSA), though she has not had a sleep study. She has been fussy and has had a poor appetite for weeks, snores heavily, and occasionally experiences brief apneic periods during sleep. She currently has a dry cough and clear rhinorrhea. The patient’s medical history is otherwise unremarkable.

➤ What tests are required as part of the preoperative evaluation for this patient?

➤ What are the anesthetic concerns for the pediatric patient with obstructive sleep apnea?

➤ What are the postoperative complications that follow tonsillectomy?

ANSWERS TO CASE 39:

Tonsillectomy/Sleep Apnea/Hemorrhage

Summary: A 2-year-old child with obstructive sleep apnea undergoes tonsillectomy under general endotracheal anesthesia. The child experiences postoperative somnolence and oxygen dependence.

➤ Preoperative evaluation: Children with tonsillar hypertrophy often have symptoms of an upper respiratory tract infection. It is important that their cough is not productive and that they are afebrile. The clinician should assess the severity of their sleep apnea by clinical indicators such as habitual snoring, apnea, daytime somnolence or hyperactivity, and lethargy. No further workup is required.

➤ Anesthesia in patients with sleep apnea: Children with obstructive sleep apnea are at higher risk for oxygen desaturation on induction and emergence from anesthesia. Moreover, the physiologic changes associated with OSA do not resolve immediately after surgery. It is prudent to consider the overnight observation of postoperative patients with significant OSA, especially those who are less than 3 years of age. Patients with OSA are also more sensitive to the respiratory depressant effects of narcotics. Great care must be taken not to administer a relative overdose while attempting to titrate to patient comfort, and naloxone should be readily available to reverse the narcosis if indicated.

➤ Postoperative complications: The most common postoperative complication after tonsillectomy is hemorrhage, which is life-threatening. The incidence of hemorrhage requiring reoperation for hemostasis is approximately 0.8%. Primary hemorrhage occurs within 24 hours of surgery, while secondary hemorrhage occurs later. Other rare and life-threatening complications include pulmonary edema, aspiration pneumonia, respiratory arrest, or massive hemorrhage.

ANALYSIS

Objectives

1. Review the preoperative evaluation of children for tonsillectomy.

2. Understand obstructive sleep apnea in children.

3. Understand the anesthetic considerations for pediatric tonsillectomy.

4. Discuss the significance of postoperative bleeding after tonsillectomy.

Considerations

This young child who presents for tonsillectomy with a diagnosis of OSA is at significant risk for perioperative complications. The ability to prepare for and deliver a safe anesthetic requires an understanding of the complex interplay of the pathophysiology of OSA, the intrinsic risks of airway surgery in a child, and the necessity of providing a balanced anesthetic. Every recent dataset on perioperative risk in pediatric patients highlights that airway surgeries are associated with the greatest incidence of adverse respiratory events. This child requires prompt support of the airway in the recovery room and preparation for invasive airway management including reintubation or continuous positive airway pressure (CPAP). Other causes for respiratory compromise should be excluded including post-obstructive pulmonary edema, pulmonary aspiration, or postoperative hemorrhage. It is unlikely that more systemic causes of arterial desaturation such as an occult cardiac lesion, intrapulmonary shunt, or pulmonary embolism are the cause. Children with OSA are susceptible to respiratory compromise after tonsillectomy and are potentially sensitive to opioids. This child most likely has postoperative respiratory depression due to the combination of chronic obstructive sleep apnea, residual anesthetic effect, and the relative narcosis caused by parenteral long-acting opioid. Other common anesthesia-related complications include postoperative nausea and vomiting (PONV), arterial desaturation, unplanned inpatient admissions, and airway obstruction.

APPROACH TO

Tonsillectomy/Sleep Apnea/Hemorrhage

DEFINITIONS

OBSTRUCTIVE SLEEP APNEA: OSA is an obstructive breathing disorder that occurs during sleep. It is characterized by intermittent upper airway obstruction due to dysfunction in the muscles and soft tissues of the upper airway, or from obstruction resulting from tonsillar hypertrophy. Most commonly, OSA is diagnosed by the history from an observer who witnesses periods of apnea. In children, common daytime symptoms of OSA include hyperactivity, failure to thrive, poor school performance, and behavioral problems.

APNEA-HYPOPNEA INDEX: AHI is the number of apneic and hypopneic episodes per hour of sleep. The AHI estimates the severity of OSA.

PERIOPERATIVE ADVERSE RESPIRATORY EVENTS: These include laryngospasm, bronchospasm, breath holding, oxygen desaturation, and severe cough.

UPPER RESPIRATORY TRACT INFECTION: URI is a common clinical condition of childhood most often caused by infection with the rhinovirus. Children contract 4 to 6 URI infections per year and require 4 to 6 weeks to return to baseline airway reactivity.

POSTOPERATIVE NAUSEA AND VOMITING: This is a common complication of anesthesia and is known to be of significant risk after tonsillectomy. PONV is of particular concern after tonsillectomy because it not only decreases patient satisfaction but it can also delay discharge and herald significant postoperative hemorrhage where blood is swallowed, unnoticed by the patient or caregiver.

CLINICAL APPROACH

Tonsillectomy is one of the most common operations performed on children. Most commonly, it is indicated on the basis of recurrent episodes of pharyngitis and obstructive sleep apnea. Tonsillectomy is increasingly performed as an outpatient procedure. While the criteria for outpatient surgical candidates continues to expand, special consideration should be given when considering outpatient surgery in patients who are less than 3 years of age, have OSA or other breathing disorders, major heart disease, bleeding diathesis, or mental retardation. Other reasons to consider inpatient admission include “acute” tonsillectomies, as well as a patient who lives far from a treatment center or who has other social circumstances that might prevent early intervention should problems develop at home. Children with proven OSA by polysomnography have been shown to have a 23% incidence of severe postoperative respiratory compromise requiring intervention. Children with Down syndrome, cerebral palsy, mental retardation, or other congenital disorders have been shown to have a 27% incidence of major respiratory compromise. In most centers, these children would not be considered as appropriate candidates for outpatient surgery.

For pediatric tonsillectomy, a thorough preoperative evaluation determines whether the child has a history of breathing disorders including sleep apnea, asthma, allergies, and any current respiratory symptoms. Patients are also screened for any history indicating a bleeding disorder. A family history of bleeding disorder or any clinical indication of bleeding disorder such as frequent and difficult to control nosebleeds or bruise formation out of proportion to injury suggests the need for additional testing prior to surgery. Routine coagulation studies are otherwise not indicated unless there is appropriate clinical suspicion. von Willebrand disease is the most common inherited coagulopathy and is not diagnosed by a screening PT/PTT.

Children who present for tonsillectomy often also manifest symptoms of an upper respiratory infection. There may be a very narrow window of opportunity for a child to be free of respiratory symptoms at the time of presentation for surgery. Some children are almost never free of symptoms. In this situation,

it is appropriate to consider proceeding with elective surgery, with the understanding that the risk of adverse respiratory complications is increased. The dependent risk factors for adverse respiratory events in children with active URIs include use of an endotracheal tube in a child less than 5 years, prematurity, reactive airway disease, parental smoking, airway surgery, copious secretions, and nasal congestion. The increase in airway reactivity may persist for several weeks after an URI.

Obstructive sleep apnea is a breathing disorder that occurs during sleep characterized by intermittent upper airway obstruction. Most commonly, the obstruction occurs at the base of the tongue or the soft palate from hypertrophic tonsils or anatomic abnormalities that narrow the aperture of the nasopharyngeal airway, allowing the tongue to be displaced posteriorly. The formal diagnosis of OSA is made by polysomnography, but for practical purposes, the diagnosis is often based on clinical symptoms. Common daytime symptoms of OSA include hyperactivity, failure to thrive, poor school performance, and behavioral problems.

Children with OSA often have hypoxemia, hypercarbia, and a partial airway obstruction even while awake. Increased airway resistance secondary to hypertrophic tonsils and adenoids can cause alveolar hypoventilation. Pulmonary artery pressure progressively increases with each apneic episode, resulting in cor pulmonale and clinically significant pulmonary hypertension, followed eventually by ventricular dysfunction. ECGs may show right ventricular hypertrophy, and some patients have x-ray findings of cardiomegaly. In most cases, surgical removal of the tonsils and adenoids can reverse these progressive cardiovascular changes.

After the preoperative evaluation, consideration may be given to premedication. Pediatric patients may often receive oral midazolam (0.3-0.5 mg/kg). However, in patients for tonsillectomy, premedication is used judiciously if there is any obstruction to breathing, and it is generally avoided in patients with severe OSA. If necessary due to high levels of anxiety, patients may receive oral midazolam (0.3-0.5 mg/kg), intramuscular ketamine (5 mg/kg), or intranasal midazolam.

Because few children would not object to placement of an intravenous, induction is performed with a standard inhalational induction using sevoflurane, nitrous oxide, and oxygen. Airway obstruction should be anticipated, and alleviated with airway manipulation, an oral airway and/or administration of CPAP at 10 to 15 cm H2O. Obese children are at increased risk of difficulty with mask ventilation. In patients with significant OSA combined with obesity or other comorbidity, an i.v. induction may be safer, albeit unpleasant for the child. An i.v. catheter allows the administration of additional anesthetic, opiates, or muscle relaxants and can thus facilitate endotracheal intubation.

An oral standard or RAE preformed endotracheal tube (ETT) is secured in the midline after placement is confirmed. However, it can become dislodged by surgical instruments, surgeon manipulation, or maneuvering of the bed.

The anesthesiologist must be vigilant to the security of the endotracheal tube at all times. A cuffed tube is preferable, since the cuff may decrease leakage of oral secretions and blood into the trachea during surgery, as well as reducing the oxygen environment at the site of electrocautery. The use of a cuffed endotracheal tube poses no additional risks in children for brief cases. Placement of the throat pack can similarly cause obstruction or kinking of the ETT. (It is also important to note and document when the throat pack is removed at the end of the case.) Maintenance is provided with standard volatile anesthetics or intravenous anesthetics such as propofol and remifentanil, which reduce postoperative nausea and vomiting as well as agitation at emergence. Muscle relaxant may or may not be necessary.

The anticipated surgical time for a tonsillectomy is 20 to 60 minutes. The surgery itself is quite stimulating, and requires a deep level of anesthetic throughout. Prior to emergence, ondansetron and dexamethasone are typically administered for vomiting prophylaxis. Metoclopramaide is no more

effective than placebo for this use. Blood and secretions must be suctioned with caution prior to emergence. Keep the suction catheter in the midline to avoid the raw surgical sites on either side. The patient should be extubated either fully awake, or (rarely) at a very deep level of anesthesia with adequate respirations after throat packs have been removed. Deep extubation may avoid the coughing and bronchospasm associated with an endotracheal tube, though larygospasm can occur should secretions or blood stimulate the vocal cords. Avoidance of Valsalva maneuvers may decrease the risk of primary hemorrhage.

Common complications of tonsillectomy include hemorrhage, postoperative nausea, vomiting, and pain (sore throat with or without otalgia). Less common complications include respiratory compromise, damage to teeth, dehydration, fever, and damage to the uvula. Rare complications include velopharyngeal insufficiency, nasopharygeal stenosis, intraoperative vascular injury, subcutaneous emphysema, mediastinistis, atlantoaxial subluxation, cervical osteomyelitis, and taste disorders. Deaths are most often due to hemorrhage, unrecognized apnea, disconnection of anesthesia circuit, and drug dosing errors or reactions, and occur at a frequency of 1:1,000-27,000 cases overall.

Hemorrhage is most likely to occur within the first 24 hours following tonsillectomy, although it is not unusual for this life-threatening complication to occur 7 to 10 days postoperatively. The reported incidence of post-tonsillectomy hemorrhage varies from approximately 0.1% to 8.1%. Anesthetic risks for a “bleeding tonsil” are significant, and include the risk of hypotension with induction reflecting the patient’s hypovolemia, and pulmonary aspiration from a stomach full of blood. Rapid-sequence induction (usually modified without the Sellick maneuver in children) should be used for post-tonsillectomy hemorrhage patients.

Without prophylaxis, the incidence of postoperative nausea and vomiting after tonsillectomy has been shown to exceed 70%. Postoperative vomiting can delay discharge, negatively impact patient and parent satisfaction, as well as increase the risk of bleeding, aspiration of gastric contents, dehydration, and electrolyte disturbances. A 2006 meta-analysis demonstrated that antiserotonergic agents, including ondansetron, granisetron, and tropisetron, are clinically effective in a dose-dependent manner. The same study provided good evidence of the efficacy of dexamethasone for vomiting prophylaxis, though the optimum dose for children remains to be determined. However, recent evidence has suggested an association between post-tonsillectomy bleeding and dexamethasone at a dose of 0.5 mg/kg.

Pulmonary edema is a rare but recognized complication of tonsillectomy. It is most likely related to the presence and relief of the airway obstruction. Excess tonsillar tissue causes an obstruction and resistance to airflow with both inspiration and expiration. This resistance results in an endogenous positive end-expiratory pressure (PEEP). Pulmonary edema seems to result from the sudden release of the excess PEEP.

Tonsillectomy is known to be a painful procedure. Traditionally, opioids have been used to treat postoperative pain, with typical doses of morphine in the range of 0.05 to 0.1 mg/kg i.v. However, opioids are known to cause respiratory depression, as well as nausea and vomiting. The respiratory depression is of special concern in patients with significant OSA, because they are likely to be more sensitive to the respiratory depressant effects of the medication. When a sleep study is available, a desaturation nadir of less than 85% predicts a significant increase of opiate sensitivity. Thus, it is recommended to reduce opiate doses by approximately 50% in these children (eg, 0.05 mg/kg).

Nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen are effective in the management of postoperative pain after tonsillectomy and do not appear to increase the risk of postoperative hemorrhage. However, surgical concern for this potentially lethal complication predisposes many against the use of NSAIDs in current clinical practice. Acetaminophen can be used concomitantly to increase the efficacy of opioids. Oral acetaminophen (15-20 mg/kg) can be administered preoperatively, or rectal acetaminophen (40 mg/kg, particularly useful in young children) can be administered after induction.

In uncomplicated patients, it is becoming increasingly acceptable to discharge patients on the same day as tonsillectomy. Pain is managed using a multimodal technique (a narcotic and acetaminophen), and dexamethasone is administered intraoperatively to reduce swelling (and thus pain) as well as postoperative nausea and vomiting. Tonsillar injection with local anesthetic may also improve analgesia. Lastly, surgical technique may affect pain after tonsillectomy. Tonsillectomy (eg, Coblation) seems to result in significantly lower pain scores postoperatively though with a possible increase in hemorrhage and recurrence rates.

Comprehension Questions

39.1. An 18-month-old child is scheduled for tonsillectomy due to recurrent pharyngitis and tonsillar hyperplasia. Preoperative evaluation should include which of the following apart from a complete history and physical?

A. Electrocardiogram

B. Polysomnography

C. Baseline hemoglobin

D. Chest x-ray

E. None of the above; no additional testing is needed

39.2. Which of the following is a symptom of obstructive sleep apnea more often seen in adults than children?

A. Irritability

B. Hyperactivity

C. Daytime somnolence

D. Depression

39.3. Regarding bleeding after tonsillectomy, which of the following statements is accurate?

A. The degree of hypovolemia is evident in the clinical presentation.

B. After 24 hours, the risk of significant bleeding becomes negligible.

C. Postoperative nausea and vomiting reduces gastric volume as well as the risk of aspiration.

D. Bleeding after tonsillectomy is a life-threatening situation.

39.4. The patient described at the beginning of this case undergoes an uneventful inhalation induction with sevoflurane. Her intraoperative course is unremarkable, and she receives 1.5 mg of morphine for pain management. She is extubated at the end of the surgery but in the recovery room, she remains somnolent and dependent on supplemental oxygen to maintain an arterial saturation greater than 92%. Which of the following is the most likely diagnosis and treatment?

A. Respiratory depression from pain, opiate

B. Low baseline oxygen saturation, PEEP

C. Cor pulmonale, ionotrope

D. Respiratory depression from excessive opiate, naloxone

ANSWERS

39.1. E. No additional testing is needed without indication from the history and physical. As described above, due to the time and financial expense of polysomnography, it is not practical to expect children to present for surgery with this data.

39.2. C. Adults with OSA will often exhibit daytime somnolence. Children are much more likely to be hyperactive and irritable, though sleep disturbance will often decrease scholastic function as well.

39.3. D. Bleeding after a tonsillectomy is a life-threatening situation. Bleeding can be quite severe, with significant hypovolemia even in a child who is not hypotensive. It is desirable that the child receives fluid resuscitation prior to induction if the situation allows. Bleeding after tonsillectomy typically occurs within hours of surgery, although a second peak in incidence occurs some 5 to 8 days later. The child with a bleeding tonsil is likely to have swallowed significant amounts of blood, warranting full-stomach precautions including a rapidsequence intubation.

39.4. D. In this case, the dose of 1.5 mg of morphine is excessive in this 14-kg child and is no doubt contributing to her postoperative narcosis. Naloxone may be indicated.

Clinical Pearls

➤ Obstructive sleep apnea is diagnosed by a child’s tendency to obstruct and remain apneic during sleep.

➤ Children with significant obstructive sleep apnea are more sensitive to opioids.

➤ The obese child with OSA undergoing tonsillectomy is at considerable risk for complications from central apnea, supraglottic obstruction, drug overdosing, and delayed emergence from volatile anesthetics.

➤ Hemorrhage is a life-threatening complication of tonsillectomy in children.

➤ The child with a bleeding tonsil can be severely hypovolemic from hemorrhage and dehydration, and is at significant risk for aspiration.

References

Bolton CM, Myles PS, Nolan T, Sterne JA. Prophylaxis of postoperative vomiting in children undergoing tonsillectomy: a systematic review and meta-analysis. Bri J Anesth. 2006;97(5):593-604. 

Brown KA, Laferriere AB, Lakheeram IM, Moss IR. Recurrent hypoxemia in children is associated with increased analgesic sensitivity to opiates. Anesthesiol. 2006;105(4):665-669. 

Ferrari LR, Vassallo SA. Anesthesia for otohinolaryngology procedures. In: Coté CJ, Todres D, Ryan JF, et al. eds. A Practice of Anesthesia for Infants and Children. 3rd ed. Philadelphia, PA: Saunders; 2001.

McColley S, April M, Carroll J, et al. Respiratory compromise after adenotonsillectomy in children with obstructive sleep apnea. Arch Otolaryngol Head Neck Surg. 1992;118:940-943. 

Randall DA, Hoffer ME. Complications on tonsillectomy and adenoidectomy. Otolaryngol–Head Neck Surg. 1998;118:61-68. 

Richmond K, Wetmore R, Baranak C. Postoperative complications following tonsillectomy and adenoidectomy—who is at risk? Int J Pediatr Otorhinolaryngol 1987;13:117-124. 

Sanders JC, King MA, Mitchell RB, Kelly JP. Perioperative complications of adenotonsillectomy in children with obstructive sleep apnea syndrome. Anesth Analg. 2006;103:1115-1121. 

Tait AR, Malviya S, Vocpel-Lewis T, Munro HM, Siewert M, Pandit UA. Risk factors for perioperative adverse respiratory events in children with upper respiratory tract infections. Anesthesiol. 2001;95:299-306.

HomeCase FilesTonsillectomy/Sleep Apnea/Hemorrhage Case File