Recommendations are followed by evidence grades (A-C) identifying the type of supporting evidence. Definitions of the evidence grades are presented at the end of the "Major Recommendations" field.
Procedural Monitoring
All patients undergoing endoscopic procedures require pre-procedural evaluation to assess their risk and to help manage problems related to pre-existing medical conditions. A history and focused physical examination, review of current medications and drug allergies, as well as an assessment of cardiopulmonary status at the time of the procedure are necessary to adequately provide for the safety of the patient. Patients undergoing endoscopic procedures should have continuous monitoring before, during and after the administration of sedatives. Monitoring may detect early signs of patient distress, such as changes in pulse, blood pressure, ventilatory status, cardiac electrical activity, and clinical and neurologic status, before clinically significant compromise occurs. Standard monitoring of sedated patients undergoing gastrointestinal endoscopic procedures includes recording the heart rate, blood pressure, respiratory rate, and oxygen saturation. Refer to the original guideline document for information on pulse oximetry, capnography, and bispectral (BIS) monitoring.
Continuous electrocardiogram (EKG) monitoring is reasonable in high-risk patients, although the necessity for such monitoring has not been shown conclusively in controlled trials. Patients who may benefit from EKG monitoring include those who have a history of significant arrhythmia or cardiac dysfunction, elderly patients, and those for whom prolonged procedures are anticipated.
Medications
The choice of sedative is largely operator dependent, but generally consists of benzodiazepines used either alone or in combination with an opiate. The most common benzodiazepines are midazolam and diazepam. The efficacy of sedation using these two benzodiazepines is comparable. However, most endoscopists favor midazolam for its fast onset of action, short duration of action, and high amnestic properties. Opiates, such as meperidine or fentanyl administered intravenously provide both analgesia and sedation. Fentanyl has a rapid onset of action and clearance and reduced incidence of nausea compared to meperidine. Morphine is usually avoided due to concern about stimulating smooth muscle contraction and inducing spasm of the sphincter of Oddi, which may be a particular problem during endoscopic retrograde cholangiopancreatography (ERCP). Combinations of benzodiazepine and opioid agents are frequently used, especially for longer procedures. However, such combinations may increase the risk of desaturation and cardiorespiratory complications. Specific antagonists of opiates (naloxone) and benzodiazepines (flumazenil) are available and should be present in every endoscopy unit to treat over-sedated patients.
Benzodiazepines
Benzodiazepines are used in the majority of endoscopic procedures. They can induce relaxation and cooperation and often provide an amnestic response. Doses are titrated to patient tolerance depending upon age, other illnesses, use of additional medications, and the level of complexity of the procedure. In addition to the desired effects, significant respiratory depression can occur. This effect is synergistically increased with the use of intravenous opiates.
Midazolam binds to stereospecific benzodiazepine receptors on the postsynaptic gamma-aminobutyric acid (GABA) neuron at several sites within the central nervous system, including the limbic system, and reticular formation. Enhancement of the inhibitory effect of gamma-aminobutyric acid on neuronal excitability results by increased neuronal membrane permeability to chloride ions. This shift of chloride ions results in hyperpolarization (a less excitable state) and stabilization. Midazolam has been found to cross the placenta and is not recommended for use during pregnancy. It also enters breast milk and is not recommended for nursing mothers.
Midazolam causes anterograde amnesia. Paradoxical reactions, including hyperactive or aggressive behavior have been reported. Midazolam has no analgesic, antidepressant, or antipsychotic properties. For dosing information on midazolam and other medications mentioned below, see the original guideline document.
Diazepam has similar properties to midazolam, although there is longer half-life and a greater chance of phlebitis and it has less amnestic capabilities. Symptoms of overdosage include respiratory depression, hypotension, coma, stupor, confusion, and apnea. Treatment for benzodiazepine overdose is supportive. Flumazenil has been shown to selectively block the binding of benzodiazepines to its receptor, resulting in reversal of central nervous system (CNS) depression. For this reason, patients who develop severe respiratory depression after being given an opioid agent and a benzodiazepine should generally receive naloxone before being given flumazenil. Acute withdrawal, including seizures, may be precipitated after administration of flumazenil to patients receiving long-term benzodiazepine therapy.
Opiates
Fentanyl binds with stereospecific receptors at many sites within the CNS and increases pain threshold, alters pain reception, and inhibits ascending pain pathways. The half-life is 2 to 4 hours. Meperidine also binds to opiate receptors in the CNS, causing inhibition of ascending pain pathways and altering the perception of pain.
Naloxone is an opioid antagonist used to reverse the sedation and respiratory depression caused by opiates. Appropriate additional supportive measures, such as fluid resuscitation and even vasopressor agents, may be required to manage cardiovascular compromise resulting from narcotic overdosage. Since naloxone may be cleared faster than meperidine, individuals who receive high doses of meperidine followed by naloxone reversal must be observed carefully to detect the development of resedation. Administration of naloxone causes the release of catecholamines, and thus should be used with caution in elderly individuals and those with cardiac diseases to avoid cardiovascular complications. In addition, administration to patients who regularly take narcotic agents may result in considerable pain and precipitate an acute withdrawal syndrome.
All narcotic agents must be used cautiously in patients taking other central nervous system depressants, such as other narcotic agents, sedatives, tranquilizers, phenothiazines, and antihistamines. Most of the drug interactions with monoamine oxidase (MAO) inhibitors have been described with meperidine. However, other narcotics should also be avoided in patients on an MAO inhibitor when possible. Narcotics will also lower the seizure threshold in patients with a history of a seizure disorder, and the dose should be lowered accordingly.
Pharyngeal Anesthesia
Pharyngeal anesthesia is often used to suppress the gag reflex during procedures involving the upper gastrointestinal tract. Commonly used topical anesthetics include benzocaine, tetracaine, and lidocaine. They are administered by aerosol spray or gargling. The effects last for up to one hour. Despite their widespread use, there are conflicting data on their benefit.
There are numerous case reports on the occurrence of methemoglobinemia after administration of topical anesthetics. This should be suspected by the presence of clinical "cyanosis" in the face of a normal arterial partial pressure of oxygen (PO2). The blood in methemoglobinemia has been variously described as dark-red, chocolate, or brownish to blue in color, and does not change with the addition of oxygen. Pulse oximetry is inaccurate in monitoring oxygen saturation in the presence of methemoglobinemia. The treatment of methemoglobinemia is with intravenous methylene blue. Routine use of topical anesthesia for upper endoscopy should be re-evaluated. It probably provides little benefit for most patients receiving the doses of intravenous sedation typically used in the United States. It may be acceptable to use topical anesthesia for some patients, particularly if light or no conscious sedation is administered.
Droperidol
Droperidol is a butyrophenone neuroleptic tranquilizer that may be used in combination with narcotics and benzodiazepines in conscious sedation for complex endoscopic procedures. It produces an anti-emetic and anti-anxiety effect. It also has mild sedative and alpha-adrenergic inhibitory action. The most common side effects are mild to moderate hypotension and tachycardia. Prolonged post-procedure drowsiness has been reported. Extrapyramidal side effects, such as dystonia, can occur.
Droperidol has been associated with QT prolongation and the development of torsades de pointes in at least 20 patients. These events have occurred in patients with no known risk factors for QT prolongation and some have been fatal. The Food and Drug Administration has issued a warning about the QT prolongation and/or torsade de pointes in patients given droperidol. This will probably greatly reduce the use of droperidol in the endoscopy setting. The drug was removed from the European market in March 2001. Droperidol should be used only in select patients with anticipated intolerance of standard sedatives and anticipated long procedure time. It should be avoided if the QTc is prolonged (>440 msec males, >450 msec females). It should be used with caution in patients at risk for development of prolonged QT syndrome, such as those with congestive heart failure (CHF), bradycardia, cardiac hypertrophy, hypokalemia, or hypomagnesemia, or if they are taking other drugs known to prolong the QT interval. Patients should remain on a cardiac monitor during the procedure and for 2 to 3 hours afterward.
Promethazine
Promethazine is an antiemetic medication which blocks postsynaptic mesolimbic dopaminergic receptors in the brain, exhibits a strong alpha-adrenergic blocking effect, and depresses the release of hypothalamic and hypophyseal hormones. It competes with histamine for the H1-receptor and reduces stimuli to the brainstem reticular system. It may be used as an adjunct to benzodiazepines and/or narcotics during endoscopy or to combat the nausea associated with these medications. Rapid administration may produce a transient fall in blood pressure. All available evidence suggests that use of occasional low doses is safe during pregnancy.
Post-procedure Monitoring
Following completion of endoscopic procedures, patients are to be observed for adverse effects from either instrumentation or sedation. The length of the follow-up observation is dependent upon the perceived risk to the patient. Patients may be discharged from the endoscopy unit or post-procedure recovery area once vital signs are stable, and the patient has reached an appropriate level of consciousness. Despite the appearance of appropriate recovery, it is well recognized that patients may have a prolonged period of amnesia and/or impaired judgment and reflexes following intravenous medications administered to induce sedation.
Patients should be advised prior to the administration of sedatives that a prolonged period of impaired cognition may occur. They should be instructed to make plans not to drive, operate heavy or potentially harmful machinery, or make legally binding decisions. When sedatives are administered, a competent companion for discharge must accompany patients from the recovery area.
Written instructions upon discharge are necessary as the amnestic period following sedation is variable. Post-procedure instruction on the signs and symptoms of potential adverse outcomes and complications is also advisable. Patients should be given written instructions on steps to follow in the event of a complication, including a phone number where 24-hour-a-day coverage is available in the event of an emergency.
Elective use of naloxone or flumazenail may be considered to reduce the recovery room time following endoscopy. The routine use of flumazenil has been shown to be associated with quicker awakening and reversal of amnesia, without an increased risk of resedation compared to placebo. Administration of antagonists following endoscopic procedures will not obviate the need for appropriate post-procedure observation and safe discharge planning. More data are needed before this becomes a recommended routine practice for outpatient endoscopy.
Special Circumstances
No Sedation
Ultrathin endoscopes with diameter from 5.3 to 6 mm can improve the tolerability of upper endoscopy. These instruments have been used through a transnasal or peroral route. They permit diagnostic evaluation of the same regions of the upper digestive tract that are accessible by standard peroral endoscopes and permit the passage of pediatric biopsy forceps to obtain tissue samples without sedation. Several studies have demonstrated comparable or improved comfort using the ultrathin upper endoscope in unsedated patients compared to standard sedated peroral upper endoscopy using standard diameter endoscopes. However, because of their small caliber, the ultrathin endoscopes are somewhat less sensitive than standard endoscopes for detecting lesions. The peroral route may be easier to perform and may be slightly better tolerated than the transnasal approach. Improvement in the equipment and increased experience with this approach will help to establish its role.
Older patients, men, patients who are not anxious, or patients without a history of abdominal pain may have better tolerance of upper endoscopy or colonoscopy with little or no sedation. For procedures performed without medications, it is still prudent to utilize varying levels of monitoring as the situation demands.
Deep Sedation or General Anesthesia
Some patients undergoing prolonged therapeutic procedures have benefited from medication such as propofol to induce deep sedation. This has been demonstrated to be superior to standard benzodiazepine/narcotic sedation for complex procedures such as endoscopic retrograde cholangiopancreatograph (ERCP). Deep sedation requires more intensive monitoring by trained individuals. Sedation-related risk factors, the depth of sedation, and the urgency of the endoscopic procedure all play important roles in determining whether or not the assistance of an anesthesiologist is needed.
Sedation-related risk factors include significant medical conditions, such as extremes of age; severe pulmonary, cardiac, renal or hepatic disease; pregnancy; the abuse of drugs or alcohol; uncooperative patients; or a potentially difficult airway for intubation. The American Society of Anesthesiologists (ASA) Taskforce states that airway management may be difficult in the following situations:
- Patients with previous problems with anesthesia or sedation
- Patients with a history of stridor, snoring, or sleep apnea
- Patients with dysmorphic facial features, such as Pierre-Robin syndrome or trisomy-21
- Patients with oral abnormalities, such as a small opening (<3 cm in an adult); edentulous; protruding incisors; loose or capped teeth; high, arched palate; macroglossia; tonsillar hypertrophy; or a non-visible uvula
- Patients with neck abnormalities, such as obesity involving the neck and facial structures, short neck, limited neck extension, decreased hyoid-mental distance (<3 cm in an adult), neck mass, cervical spine disease or trauma, tracheal deviation, or advanced rheumatoid arthritis
- Patients with jaw abnormalities, such as micrognathia, retrognathia, trismus, or significant malocclusion.
The ASA Taskforce guidelines recommend that the presence of one or more sedation-related risk factors, coupled with the potential for deep sedation, will increase the likelihood of adverse, sedation-related events. In this situation, if the practitioner is not trained in the rescue of patients from general anesthesia, then an anesthesiologist should be consulted. The routine assistance of an anesthesiologist for average risk patients undergoing standard upper and lower endoscopic procedures is not warranted and is cost-prohibitive.
Summary
- A focused history and physical is required prior to the administartion of moderate sedation (C).
- Routine monitoring of the patient's pulse rate, blood pressure, oxygen saturation, are useful identifying early problems (B). Monitoring of EKG recordings may be helpful in selected cases (C). Capnography, measurement of carbon dioxide retention, may be useful in prolonged cases (A).
- The use of benzodiazepines and/or opiates will result in a satisfactory outcome in nearly all patients (B). Endoscopists prefer the combination of these drugs, but it adds little benefit from the patient's viewpoint (A).
- Specific antagonists of opiates (naloxone) and benzodiazepines (flumazenil) are available and should be present in every endoscopy unit to treat over-sedated patients (C).
Definitions:
- Randomized controlled trials
- Non-randomized studies
- Expert opinion
