Endoscopy Practice and Sedation: Sedation Monitoring and Special Populations

0:00Welcome to Board Pearls. This is episode three of four of the Endoscopy Practice and Sedation chapter, in the Endoscopic Procedures module. In this episode we cover the safety apparatus that sits on top of the sedation agents: capnography and airway prediction, the naloxone and flumazenil reversals, the special-population modifications including cirrhosis, obstructive sleep apnea, the elderly, and pregnancy, and the updated GLP-1 periprocedural guidance of twenty-four hours of clear liquids.

0:29The last episode was about matching the drug to the gap between the depth you want and the depth you can rescue from. This episode is about the monitoring, prediction, and rescue that catch the patient when the drug carries them past the intended depth, and then the population-specific tweaks that reshape the whole plan when one shift in pharmacokinetics or aspiration risk changes what safe looks like.

0:51Start with monitoring, because the safety triad of capnography, airway prediction, and reversal carries most of the testable material. Capnography is the recommended standard at moderate and deep sedation depths and the reason is mechanistic. End-tidal CO2 reflects alveolar ventilation in real time, within one to two breaths. Pulse oximetry reflects arterial saturation, which lags hypoventilation by thirty to ninety seconds in a patient on supplemental oxygen, because the oxygen reservoir delays desaturation. Loss of the capnographic waveform is therefore the earliest reliable sign of apnea. Supplemental oxygen actively masks the desaturation that would otherwise alert the team. The boards test the temporal lag directly: a sedated patient on nasal cannula whose saturation reads ninety-eight while the capnograph waveform has gone flat is apneic, not stable. Capnography reduces hypoxia and rescue events in randomized trials and is the standard of care.

1:51Airway prediction tells you whether the moderate-sedation plan is dangerous before the case starts. The Mallampati classification grades oropharyngeal exposure with the mouth open and tongue protruded. Class one shows the full soft palate, uvula, and pillars. Class two shows the soft palate and the base of the uvula. Class three shows only the soft palate. Class four shows only the hard palate. Class three and four predict difficult mask ventilation and difficult intubation. The other features that travel with difficult airways are worth holding together. Short thyromental distance under six centimeters, limited cervical mobility, prior difficult intubation, micrognathia or retrognathia, large tongue, short neck, advanced rheumatoid arthritis with cervical involvement, trisomy twenty-one, or BMI above thirty-five. Any combination warrants anesthesia consultation before deep sedation is offered. The patient with a Mallampati four and a BMI of forty-four and severe sleep apnea is not getting propofol from the endoscopist. That patient is getting propofol from anesthesia or general anesthesia with a secured airway.

3:03Aspiration risk modifies depth and technique. The drivers are the same physiology that defined the fasting exceptions earlier in the chapter. Gastroparesis, achalasia, gastric outlet obstruction, recent meals, retained gastric contents on a GLP-1 receptor agonist, advanced pregnancy after the first trimester, and a full bowel preparation in a colonoscopy patient with delayed transit. Any of these warrants either rapid sequence induction with a cuffed endotracheal tube, or a procedural delay until the risk is mitigated. Rapid sequence induction minimizes the window between loss of consciousness and intubation. Cricoid pressure, the Sellick maneuver, is commonly applied during the induction to externally compress the esophagus, though the evidence behind it is debated.

3:49Reversal closes the loop. Flumazenil reverses benzodiazepines by competitive antagonism at the GABA-A benzodiazepine binding site. The dose is zero point two milligrams intravenous over fifteen seconds, repeated every minute up to a cumulative one milligram. Onset is within one to two minutes. The contraindications are predictable from the mechanism. Chronic benzodiazepine use means abrupt antagonism precipitates withdrawal seizures. A seizure disorder means the same risk on a different physiologic substrate. In a mixed overdose with tricyclic antidepressants, flumazenil can unmask the TCA-induced seizure that the benzodiazepine effect had been suppressing. The other piece worth holding is the duration mismatch. Flumazenil lasts thirty to sixty minutes. Midazolam can outlast it. Resedation is therefore expected, and post-reversal monitoring runs for at least two hours.

4:45Naloxone reverses opioids by competitive antagonism at the mu-opioid receptor. The procedural sedation dose is zero point zero four to zero point four milligrams intravenous every two to three minutes, titrated to respiratory rate and consciousness rather than to full arousal. The goal is restoration of breathing, not awake-and-fighting. Abrupt full reversal in an opioid-tolerant patient produces acute pain, hypertension, tachycardia, and pulmonary edema, which is why the titration is graded. The larger zero point four to two milligram bolus is reserved for unwitnessed overdose. Duration is thirty to ninety minutes, shorter than most of the opioids it reverses, and much shorter than methadone or sustained-release formulations. Monitoring for at least four hours after reversal is the rule, longer for the long-acting agents.

5:35That is the basic apparatus. The special-population modifications run the same logic through a different physiology, and the value of working through them is that one shift in pharmacokinetics or aspiration risk reshapes the whole plan.

5:50Cirrhosis is the canonical case. Reduced CYP3A4 activity prolongs midazolam, and active midazolam metabolites accumulate when hepatorenal syndrome adds renal impairment. The result is that the same one milligram bolus produces a deeper and longer effect than the same dose in a non-cirrhotic patient. The clinical move is to halve the midazolam dose and extend the titration interval, and many centers avoid benzodiazepines entirely in cirrhosis to preserve mental status. Fentanyl is similarly prolonged through reduced hepatic clearance, but it is preferred over morphine because the metabolites are inactive. Propofol clearance is less affected because hepatic blood flow rather than enzymatic capacity drives it, which is the reason propofol becomes the preferred sedative in cirrhotics requiring sedation. The trade is hemodynamic. Cirrhotic patients are often hypovolemic and propofol's vasodilatory effect produces more pronounced hypotension than usual. The bigger concern is encephalopathy. The cirrhotic central nervous system is sensitized to GABAergic agents and ammonia-driven encephalopathy is unmasked by sedation depth. The Child class B patient with a history of overt encephalopathy is the patient where benzodiazepines are avoided entirely, and propofol delivered by anesthesia is the plan.

7:07Elderly patients, over sixty-five, take a twenty-five to fifty percent dose reduction across all sedation agents. The drivers converge from multiple directions. Reduced volume of distribution. Reduced hepatic metabolism. Reduced renal clearance. Increased CNS sensitivity from age-related changes in GABA receptor density and function and reduced cerebral blood flow. Peak effect is delayed, three to seven minutes for midazolam in the older patient versus three to five in the younger adult, which means titration intervals lengthen. Falls and prolonged sedation are the dominant adverse outcomes, and standard institutional fall precautions apply through recovery.

7:45Obstructive sleep apnea increases pharyngeal collapsibility and softens the threshold for respiratory depression. Capnography is essential. Continuous positive airway pressure during recovery helps in patients with a home CPAP. Deeper sedation with airway-trained personnel is preferred over moderate sedation that may drift into deeper sedation without the personnel to catch it. Many centers use propofol-MAC routinely for OSA patients to keep airway control deliberate rather than accidental. STOP-BANG screening before the procedure identifies the undiagnosed OSA patient. The acronym runs snoring, tiredness, observed apnea, blood pressure, BMI, age, neck circumference, and gender. A patient with three or more positive features carries enough probability of clinically significant OSA to change the sedation plan.

8:40Pregnancy beyond the first trimester delays gastric emptying through progesterone-mediated smooth-muscle relaxation and mechanical compression by the gravid uterus. That combination is why pregnant patients beyond the first trimester are managed with full-stomach precautions. Left-lateral positioning during the procedure prevents aortocaval compression. Fetal monitoring is appropriate after twenty-two to twenty-four weeks when fetal viability and intervention become possible. The agent choices follow placental crossing and FDA category. Propofol is category B and is preferred when sedation is required, across all trimesters including the first. Midazolam is category D and is avoided particularly in the first trimester because of historical cleft lip and palate concerns. Fentanyl is category C and is acceptable in moderate doses. Ketamine is category B but produces uterine hypertonus and is avoided. Elective endoscopy is generally deferred to the second trimester or postpartum when the indication permits, because the first trimester carries teratogenicity risk and the third trimester carries higher risk of preterm labor.

9:46That leaves the newer drug class that has rewritten the periprocedural fasting conversation. GLP-1 receptor agonists, semaglutide, tirzepatide, liraglutide, dulaglutide, delay gastric emptying through GLP-1 receptors on gastric smooth muscle and through central appetite suppression. The clinical consequence is retained gastric content at the time of the procedure even in a patient who fasted normally, and aspiration when that content regurgitates into an unprotected airway under sedation.

10:21The October twenty twenty-four multisociety guidance pivoted away from a uniform hold. The earlier reflex was to stop the agent for one dosing interval before any endoscopy, a week for the weekly drugs, a day for the daily ones. The current framework is risk-stratified and the headline recommendation is different. Elevated aspiration risk is defined by a few features. Active GI symptoms such as nausea, abdominal distention, or fullness. The dose escalation phase. Weekly dosing regimens that produce sustained receptor activation. For these patients, the recommendation is a clear liquid diet for at least twenty-four hours before the procedure while continuing the GLP-1 agonist. Holding the agent for one dosing interval is now the fallback, reserved for cases where shared decision-making concludes the medication should be held.

11:10For patients without elevated risk who present on the day of the procedure without having held the medication, three options are on the table. Point-of-care gastric ultrasound to evaluate for retained content, if available and the operator is proficient. Treating the patient as full-stomach with rapid sequence induction and a cuffed endotracheal tube. Or proceeding cautiously under shared decision-making. And one explicit carve-out: time-sensitive endoscopy should not be delayed for GLP-1 management. The urgent diagnostic procedure, the urgent therapeutic intervention for bleeding or obstruction, goes forward, because the marginal aspiration risk from continued GLP-1 use is smaller than the harm from delaying a clinically necessary procedure. The clinical move in that setting is to plan the airway aggressively, not to reschedule.

11:57Pull the safety apparatus together. Capnography catches apnea before pulse oximetry desaturates, which is why it is the standard at moderate and deep depths regardless of who is running the case. Mallampati, thyromental distance, and the rest of the airway predictors decide whether the standard plan is safe or whether anesthesia comes in before the first dose. Flumazenil and naloxone reverse benzodiazepines and opioids respectively, but both wear off faster than the agents they reverse, which is why monitoring runs hours past the reversal dose.

12:29Special populations are not exceptions to the framework. They are the framework applied through a shifted pharmacokinetic and aspiration-risk profile. Cirrhosis halves the benzodiazepine dose and shifts toward propofol delivered by anesthesia. Elderly patients take a quarter to a half off across the board with longer intervals between boluses. OSA patients get capnography and deeper sedation with airway-trained personnel rather than a moderate plan that may drift. Pregnancy beyond the first trimester gets full-stomach management with propofol as the preferred agent. And the GLP-1 patient is now managed with twenty-four hours of clear liquids on the agent rather than a reflex hold, with full-stomach precautions or gastric ultrasound on procedure day when the medication was not adjusted in advance.

13:14The next episode closes the chapter on the topics that sit alongside sedation but run on different logic. Antibiotic prophylaxis through the closed-space-infection principle, with the misconception that prosthetic devices change the indications. Pacemakers and ICDs and the electromagnetic interference of monopolar electrosurgery, with the role of bipolar containment and magnet placement. Capsule endoscopy contraindications and the patency-capsule workup for retention risk. And the ASGE adverse event lexicon that grades severity by intervention intensity across mild, moderate, severe, and fatal. The question that opens the next episode is what changes about the procedure plan when the patient brings hardware, prosthetic anatomy, or a closed space the immune system cannot reach.

14:03For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode three of four of chapter twenty eight, and I'll see you in the next one.