Anesthesia for Upper Gastrointestinal Endoscopy

Patient Populations Presenting for Upper Gastrointestinal Endoscopic Procedures

Gastroesophageal reflux disease and its potential sequelae, including Barrett’s esophagus and esophageal cancer, are common indications for EGD. The incidences of Barrett’s esophagus and esophageal cancer are increasing. Barrett’s esophagus is a precursor of esophageal cancer. Fortunately, effective endoscopic therapeutic interventions are available that can decrease the need for a debilitating esophagectomy. However, the endoscopic interventions can be lengthy and uncomfortable. Many patients who undergo esophagectomy enjoy long-term survival, but they frequently present to the gastroenterologist for repeated dilations.

Conditions for which surgery was seen as the only curative option, such as tracheoesophageal fistula, are now treated endoscopically with a high degree of success. Bariatric surgery is gaining popularity in developed nations, and EGD is often performed for assessment before or after the surgery. Placement of stents, cautery of bleeding arteriovenous malformations, and banding of esophageal varices are other growing areas of endoscopic practice.

The most frequent indication for ERCP is common bile duct stones. Gallstone disease is estimated to affect more than 20 million adults in the United States, at an annual cost of $6.2 billion. Of patients undergoing laparoscopic cholecystectomy for symptomatic cholelithiasis and patients with acute biliary pancreatitis, 5% to 10% and 18% to 33%, respectively, also will have choledocholithiasis.

Patients with pancreatic cancer often need endoscopic ultrasound or ERCP for staging or palliation. The pancreas is the 10th most common site of new cancer and the 4th-leading cause of cancer death in the United States and worldwide. In 2010 an estimated 43,140 new cases of pancreatic cancer were diagnosed in the United States; this accounts for 2.8% of all cancers.

Although many advanced endoscopic procedures are currently the privilege of tertiary gastroenterology centers, they are likely to be performed more widely in the future.

Preprocedure Evaluation

In general, patients presenting for an inpatient EGD or ERCP are likely to be far sicker than their outpatient counterparts. The comorbidities discussed in the following paragraphs should be kept in mind while conducting a preoperative evaluation. The time available for a comprehensive evaluation in a busy endoscopy unit may be limited; the anesthesiologist should triage and allocate appropriate time for a given patient.

Hypertension and coronary artery disease are extremely common in this patient population. The procedure in a patient presenting for surveillance EGD who has undergone recent coronary stent placement, with either a bare metal or drug-eluting stent, may be postponed if issues concerning ongoing anticoagulation are not addressed. If the patient presents for cancer staging or for a therapeutic ERCP, however, the risks of postponement might exceed the risks of the procedure. The risk-to-benefit ratio for fine-needle aspiration cytology and any other potential instigation of bleeding should be discussed in advance with the gastroenterologist. A cardiologist should also be consulted for guidance in these situations.

Cardiac rhythm disorders are not uncommon. Chronic atrial fibrillation with anticoagulation is less frequent, with wider availability and success of antiarrhythmic medications and ablation procedures. Patients with a pacemaker or automatic implantable cardioverter-defibrillator are increasingly encountered. Therapeutic EGD or ERCP might involve use of electrocautery close to these devices. It is prudent to have a magnet available, although prophylactic use is unnecessary. When applied, the magnet may activate the backup mode (fixed-rate pacing); in such an event, pacemaker evaluation and resetting are recommended.

Ventricular assist devices (VADs) are more commonly used. In our hospital, these patients are managed by noncardiac anesthesiologists for noncardiac procedures. In the absence of pulsatile flow, pulse oximeter and noninvasive blood pressure readings may be unreliable. Communication with the VAD coordinator nurse regarding physiological parameters such as volume and vasodilation status is important.

Patients with chronic obstructive airway disease who require supplemental oxygen at baseline may occasionally present for EGD or ERCP. These patients usually can be anesthetized for endoscopy without endotracheal intubation and sent home on the same day.

Obstructive sleep apnea is present in approximately 70% of patients with morbid obesity presenting for bariatric surgery. An accurate airway assessment is important. Intubation is rarely necessary in these patients for EGD, although elective intubation may be considered for prolonged ERCP.

Considering the frequency of diabetes mellitus in the population, it is impossible to schedule all such patients as the first case of the day. Most of these patients can be fed almost immediately after the procedure and return to their insulin or oral treatment regimen. Holding the morning dose of hypoglycemic agent and insulin may be all that is necessary.

Patients on chronic opioid therapy require significantly higher than average doses of intravenous anesthetic agents and opioids. Gastroenterologists will often request anesthesia for these patients.

Endoscopic procedures have low cardiac risk. Therefore, according to recent guidelines, preprocedural electrocardiograms (ECGs) are not indicated in asymptomatic patients, even if they are older than 60 years of age or have known cardiovascular disease. Likewise, laboratory testing is not usually needed before endoscopy, with rare exceptions such as coagulation studies and platelet count in patients with liver disease undergoing therapeutic endoscopic procedures.

Anesthetic Drugs for Gastrointestinal Endoscopy

Propofol is the cornerstone when anesthesia services are provided for EGD and ERCP. Smooth intravenous induction and maintenance, quick and clear-headed recovery, and low incidence of postprocedure nausea and vomiting make it an attractive anesthetic agent in this setting. The anesthesia provider should, however, be mindful of the fact that patients are often under general anesthesia during some or most of an endoscopy procedure with propofol; this has attendant risks, especially loss of airway control. The anesthesia provider must also appreciate the pharmacokinetic and pharmacodynamic variability of propofol, especially when used in combination with other drugs (such as short-acting opioids). As a result, a patient with a seemingly adequate dose can be still awake, and a patient with a smaller dose can be apneic, be unresponsive, and require airway support. Additionally, instrumentation of an inadequately sedated patient can cause reflex coughing and laryngospasm despite the fact that propofol does not increase airway irritability.

Although propofol is not absolutely contraindicated in patients with a mild egg allergy, patients occasionally present with a history of severe egg allergy or propofol allergy. One alternative to propofol is etomidate. Etomidate is advantageous because of its minimal cardiorespiratory depression but disadvantageous because of the increased incidence of nausea or vomiting and tonic or clonic movements in contrast to propofol. Both etomidate and propofol may produce a burning sensation when injected intravenously; this unpleasant effect can be mitigated by pretreatment with intravenous lidocaine, which may also have some general anesthetic properties and suppress coughing from endoscope insertion.

Another alternative to propofol is the combination of a benzodiazepine and an opioid such as midazolam and remifentanil. Remifentanil is an ultra–short-acting opioid. Its pharmacodynamic properties are similar to those of the other commonly used opioids, but its pharmacokinetic properties are unique. Unlike all other opioids, remifentanil is metabolized by esterases, because of the presence of an ester moiety at the active site, and esterases are ubiquitous, being present in many tissues, including plasma and red blood cells. Its fixed context-sensitive half-time stems from this unique metabolism. As a result of rapid elimination, its context-sensitive half-time is approximately 2 to 4 minutes. It is important to administer it as a small bolus (1 to 2 mcg/kg) infused over a period of at least 2 minutes. This is followed by an infusion at 0.06 to 0.2 mcg/kg/min. Administration of 0.5 to 2 mg midazolam 2 or 3 minutes before is useful for amnesia, although remifentanil itself has amnestic properties. Midazolam can be repeated again in small doses during the procedure. After 4 to 5 minutes of remifentanil infusion, patients tolerate GI endoscopy quite well. It is important not to be tempted to increase the remifentanil infusion, because apnea and rigidity can ensue. Frequently, patients receiving remifentanil can stop breathing, although they are conscious and cooperative, and they need to be reminded to take deep breaths. Supplemental oxygen should be administered. The patients usually tolerate endoscopy well and often have no recollection of the procedure.

Limited data exist regarding the use of ketamine for endoscopy. Ketamine produces a less familiar state of anesthesia: the patient might give an impression of staring, with eyeballs rolling, and he or she might even move. The mere mention of ketamine may conjure up concern about psychomimetic reactions, but the incidence of those can be greatly reduced by coadministration of midazolam, propofol, or both. Although ketamine’s cardiovascular stimulation may be undesirable in some patients (such as those prone to myocardial ischemia), it may be advantageous in some hemodynamically unstable (e.g., septic or bleeding) patients. In most patients, ketamine will offset hypotension caused by propofol. One potential problem with ketamine for endoscopy is that it promotes salivation, which can be mitigated by preprocedure anticholinergic administration but that is associated with further tachycardia and postprocedural dry mouth. Ketamine’s lack of respiratory depression is attractive for upper GI endoscopy, particularly in patients with obstructive sleep apnea. Ketamine may also be particularly advantageous in patients with asthma because of its bronchodilating properties. Ketamine is an analgesic as well, so additional analgesics may not be needed in patients undergoing endoscopy.

Dexmedetomidine, a selective α ₂ -adrenergic receptor agonist, has properties unlike those of other commonly used anesthetic drugs. Its action is similar to that of clonidine, another α ₂ agonist, but it is six times more selective. The drug is now approved for sedation in nonintubated patients, so its inclusion in the GI endoscopy armory is relevant. Almost all studies agree that patients receiving dexmedetomidine maintain spontaneous ventilation with almost no assistance, apart from an occasional chin lift. A few studies have been conducted on the role of dexmedetomidine in GI endoscopy. Coadministration of midazolam and a short-acting opioid will help achieve a higher success rate. Dexmedetomidine is administered with a loading dose over 10 minutes, and its effects of sedation and hypotension tend to linger longer than those of propofol. Thus, although dexmedetomidine may be appropriate for certain patients, such as those particularly prone or sensitive to respiratory depression from other agents, it is not as practical for routine, rapid turnover cases.

In summary ( Table 14-1 ), for most patients, propofol is the basis of anesthesia for upper GI endoscopy although in particular patients alternative drugs may be more attractive. The details of intravenous administration, especially target-controlled infusion (TCI) versus traditional intravenous infusion, are discussed in a separate chapter, so discussion here is limited to their application for endoscopy. Depending on which side of the Atlantic Ocean one is practicing, the tools for intravenous anesthetic delivery will vary. Use of TCI pumps will definitely reduce the variability, and that reduction might be especially useful for endoscopy. They are known to reduce the number of changes to the infusion settings necessary for surgical anesthesia. In the setting of endoscopy, however, in which the change of stimulation is less frequent than in surgical procedures, such a luxury may not be necessary.

Airway Management

Anesthetizing in any location needs to be taken as seriously as in an operating room. Preoperative evaluation with special attention to airway and aspiration risk factors must be thorough. An anesthesia machine is not required, but a breathing circuit capable of positive-pressure ventilation (such as Mapleson C), laryngoscope, face masks, oral and nasal airways, laryngeal mask airways, various sizes of endotracheal tubes, and emergency drugs should be readily available. Because of its remote location, it is also important to have additional airway adjuncts such as Bougies, stylets, video laryngoscopes, and carbon dioxide detectors. It is important to check the availability and functionality of these before the start of every procedure. Airway emergencies such as laryngospasm and other airway obstruction often occur with little notice during upper GI endoscopy. Being ready for any airway situation (or not) could be the difference between apnea-related cardiac arrest and a safe discharge home.

Options for airway management for upper GI endoscopic procedures range from natural (with supplemental oxygen by nasal cannula) to endotracheal intubation. In fact, a study of ERCPs with laryngeal mask airways has been published. Another choice used in our practice is insertion of a nasal trumpet. Insertion of a nasal airway may provoke epistaxis, so use of a vasoconstrictor nasal spray or oral insertion of the nasal trumpet should be considered, particularly in patients who have received anticoagulation therapy. Even for advanced endoscopic procedures, including complicated ERCPs, endotracheal intubation is rarely used in our practice, with the exception of patients with a stomach full of food or blood. It should be noted that gastric contents can be suctioned by the endoscopist under direct vision.

The experience of the anesthesiologist in providing anesthesia for endoscopy procedures seems to play a major part in this decision; anesthesiologists unfamiliar with this area of practice seem to intubate more frequently. It is a tightrope walk to keep the patients unresponsive and comfortable while spontaneously breathing. Avoiding endotracheal intubation is beneficial in the rapid turnover, outpatient endoscopy setting. In this population, irrespective of the drugs chosen and the tool used to administer them, the central aim of anesthesia is to have an unresponsive patient who is spontaneously ventilating and maintaining the airway with little or no support. Airway manipulations such as chin lift, jaw thrust, and neck extension and various devices help maintain a patent airway, and their use should not be seen as failure of the technique. Having a dedicated team of providers with experience in anesthesia for endoscopy and an appreciation of pharmacokinetic and pharmacodynamic variability will help master this technique and greatly facilitate efficiency without compromising safety. The gastroenterologist must be aware that scope removal during the procedure may be needed urgently for airway management. Not every situation, procedure, and patient factor can be addressed, but the following items can be used as broad principles for airway management in upper GI endoscopy.

• 1.

  Most patients with normal airway anatomy and physiology presenting for a short diagnostic upper GI endoscopy do not need any special airway apart from nasal cannula for supplemental oxygen administration.

• 2.

  Most ERCPs in our hospital are done without an endotracheal tube. Anesthesia is induced after prone positioning. A nasal trumpet can be inserted soon after induction and connected to a breathing system, as seen in Figure 14-1 . Apnea lasting 30 to 45 seconds after induction is not uncommon, but the stimulation of gastroscope insertion helps restart spontaneous ventilation. The nasal trumpet allows some degree of controlled ventilation, if necessary. More importantly, it allows delivery of 100% oxygen at the laryngeal inlet. High-frequency jet ventilation is also useful in this setting. Endotracheal intubation is the airway of choice for endoscopic drainage of a pancreatic pseudocyst.

  

• 3.

  Most therapeutic upper GI endoscopies such as endoscopic mucosal resection, Barrett’s eradication, application of variceal banding, and resection of larger gastroduodenal polyps are managed similarly to ERCPs. These procedures may involve frequent scope withdrawal and reinsertion, so it is important to maintain or increase the depth of anesthesia to prevent coughing. Patients undergoing treatment for tracheoesophageal fistula, especially with a history of aspiration, should be considered for endotracheal intubation.

• 4.

  Morbidly obese patients often present for upper GI endoscopy before weight reduction surgery. Obstructive sleep apnea is very common in this group. We use nasal trumpets after induction, occasionally supplemented with supraglottic jet ventilation. Use of supraglottic jets provides ventilation and also probably prevents the upper airway from collapsing. This is a technique that requires experience and maintenance of adequate depth of anesthesia at all times. If in doubt, these patients are better handled with endotracheal intubation. In any event, equipment for difficult intubation should be at hand.

• 5.

  Patients who had previous esophagectomy for cancer or achalasia frequently present for esophageal dilation. In the absence of gastric motility issues, these patients can be handled safely with supplemental nasal oxygen or a nasal trumpet connected to breathing system. Stretching with a balloon or a Bougie can be stimulating, and deepening of anesthesia in anticipation is important.

• 6.

  Patients with a documented pharyngeal pouch are anesthetized after awake endotracheal intubation. Application of cricoid pressure may not be useful in this scenario.

• 7.

  Patients with limited mouth opening or neck extension as a result of radiation treatment for oropharyngeal cancer often present for repeated esophageal dilations of radiation-induced stricture. In the absence of any nasopharyngeal airway obstruction and if ventilation by mask is not expected to be difficult, these patients can be safely managed with nasal cannula or nasal trumpet.

• 8.

  Patients who had prior weight loss surgery sometimes present for endoscopic evaluation. Insufficient evidence exists to recommend endotracheal intubation in this subgroup. Each individual patient needs to be evaluated with regard to potential for aspiration and accordingly managed.

• 9.

  GI bleeding is common in patients with ventricular assist devices. Although it is recommended to treat them as having a full stomach because of the position of the devices, in our practice oxygen is delivered with nasal cannula or nasal trumpet for EGD. The potential risks of rapid sequence induction and intubation in these very sick patients outweigh any benefits.