Volume 35, No. 2 • June 2020   Issue PDF

Gas Embolism Events in GI Endoscopy: What We Know Now and How We Can Mitigate Them Through the Routine Use of CO2 Insufflating Gas Instead of Air

Nikolaus Gravenstein, MD; Brian Thomas, JD

There is an accumulating pulmonary and/or cardiac adverse event case experience arising out of gastroenterology (GI) endoscopy suites. High morbidity and mortality embolic events have been reported when the insufflating gas from a gastroenterologist’s endoscope enters a patient’s circulation.1 It is those gas embolism events that are often conflated with poor anesthetic care, and where there is still, in many cases, a demonstrable opportunity to reduce harm by using carbon dioxide instead of air as the endoscope insufflating gas. We are reminded of this in a recent infographic.2

Anesthesia professionals are involved in endoscopy suite cases to enable patient cooperation, immobility, and amnesia. These cases are variably described as monitored anesthesia care or general anesthesia. We think they are best described as intravenous general anesthetics (drug-induced loss of consciousness during which patients do not arouse during otherwise painful stimulation), i.e., the patient accepts endoscope insertion and manipulation. If one accepts the target patient state as one of at least moderate sedation ranging to general anesthesia, expected monitoring includes capnography. Capnography in the setting of what is often a natural, and in the case of upper endoscopy procedures also a shared, airway is technically more challenging than in an endotracheally intubated patient. Nasal cannula and/or oral airway derived capnography makes it difficult to reliably and quickly identify clinically significant decreases in ETCO2, as would occur with a gas embolism. In an intubated patient a drop in ETCO2 is much more obvious and unambiguous because there is generally a very stable ventilation and gas sampling relationship with every exhaled breath. The vagaries of natural airway capnography may obfuscate an otherwise significant drop in ETCO2. This might delay provider recognition of perturbations in pulmonary blood flow, blood pressure, heart rate, or oxygen saturation that are associated with significant gas embolism.

Clinically significant gas embolization during an endoscopy procedure can happen any time there is a connection between the insufflating gas and the vascular system and intraluminal pressure is higher than local vascular pressures. When the venous pressures are generally less than 20 mmHg, the natural question to ask is: Can the gas pressure at the tip of a GI endoscope be higher than venous pressure? The answer is, perhaps surprisingly, very much so. In an in vitro model, it was determined that the gas pressure at the tip of the endoscope may easily exceed 175 or even 300 mmHg depending on the endoscopy system and flow settings.3 This is possible because the GI endoscopy system insufflating gas is flow- and not pressure-limited. Thus, if there is no place for the insufflating gas to decompress around the scope, then the local venous pressure is easily exceeded by the gas pressure which, along with a mucosal or vascular breach (e.g., biopsy, myotomy, ulcer, inflammation, necrosectomy, dilation, or a stent placement), enables a gas embolism scenario.

Not all endoscopy procedures have the same risk. In purely diagnostic procedures, the risk is indeed negligible. There is a strong argument in favor of making CO2 the default insufflating gas for all GI endoscopy procedures, because we cannot consistently predict when a biopsy will be taken, a mucosal surface to be breached, and the insufflating gas pressure at the tip of the endoscope also exceed venous pressure. Carbon dioxide insufflated laparoscopic surgeries, set a safety precedent for this which we are all familiar with. This is in part, because CO2 is more readily absorbed and less likely to be lethal than air if it gets into the vascular system.

Using carbon dioxide for GI endoscopy is not novel. In 1974 it was suggested to use carbon dioxide to reduce the risk of explosion with colon polyp cautery.4 Yet, in a 2008 survey, fewer than half of the endoscopists reported that they were aware that using CO2 for insufflation gas was an option for GI endoscopy procedures and <5% used CO2 for insufflation.5 Another potential benefit of the CO2 gas use is that it may cause less postprocedure discomfort than air.6

We as anesthesia professionals should support the use of CO2 insufflating gas during GI endoscopic procedures. It is readily available and demonstrably safer.7 To help persuade yourselves and gastroenterology colleagues, the 2016 American Society for Gastroenterology Technology Committee reported that, “Several authors recommend the use of CO2 instead of room air as an insufflation agent during endoscopy because of the rapid tissue absorption of CO2, in the event that gas embolism takes place. This recommendation appears to be particularly valid for higher risk interventions including ERCP, cholangioscopy, and endoscopic necrosectomy.”8


Dr. Gravenstein is professor of Anesthesiology, Neurosurgery and Periodontology at the University of Florida.

Brian Thomas, JD, is vice president—Risk Management for Preferred Physicians Medical, a medical professional liability insurance carrier that provides malpractice insurance exclusively to anesthesiologists and their practices.

The authors have no conflicts of interest.


  1. Donepudi S, Chavalitdhamrong D, Pu L, et al. Air embolism complicating gastrointestinal endoscopy: a systematic review. World J Gastrointest Endosc. 2013;5:359–65.
  2. Wanderer JP, Nathan N. Bubble trouble: venous air embolism in endoscopic retrograde cholangiopancreatography. Anesth Analg. 2018; 127:324.
  3. Bursian A, Gravenstein N, Draganov PV, et al. GI endoscopy insufflating gas pressure: how regulated is the regulator? Paper presented at: International Anesthesia Research Society Annual Meeting; May 2016; San Francisco, CA.
  4. Rogers BH. The safety of carbon dioxide insufflation during colonoscopic electrosurgical polypectomy. Gastrointest Endosc. 1974;20:115–117.
  5. Janssens F, Deviere J, Eisendrath P, et al. Carbon dioxide for gut distension during digestive endoscopy: technique and practice survey. World J Gastroenterol. 2009;15:1475–1479.
  6. Wang WL, Wu ZH, Sun Q, et. al. Meta-analysis: the use of carbon dioxide insufflation vs. room air insufflation for gastrointestinal endoscopy. Aliment Pharmacol Ther. 2012;35: 1145–1154.
  7. Afreen LK, Bryant AS, Nakayama T, et al. Incidence of venous air embolism during endoscopic retrograde cholangiopancreatography. Anesth Analg. 2018;127:420–423.
  8. ASGE Technology Committee, Lo SK, Fujii-Lau LL, Enestvedt BK, et al. The use of carbon dioxide in gastrointestinal endoscopy. Gastrointest Endosc. 2016;83:857–865.

Additional references:

Sharma VK, Nguyen CC, Crowell MD, et al. A national study of cardio- pulmonary unplanned events after GI endoscopy. Gastrointest Endosc. 2007;66:27–34.

Ben-Menachem T, Decker GA, Early DS, et al. Adverse events of upper GI endoscopy. Gastrointest Endosc. 2012;76:

Ali Z, Bolster F, Goldberg E, et al. Systemic air embolism complicating upper gastrointestinal endoscopy: a case report with post-mortem CT scan findings and review of literature. Forensic Sci Res. 2016;1:52–57.

Dellon ES, Hawk JS, Grimm IS, et al. The use of carbon dioxide for insufflation during GI endoscopy: a systematic review. Gastrointest Endosc. 2009;69:843–849.