Feedback on our Apiration Discussion on "Fortnight"

John and I have appreciated all the feedback we’ve gotten on “This Fortnight in Medicine.” It’s been fun to dive into a couple of articles every other week, and the emails and comments from listeners have made it even better.

Our discussion of the recent meta-analysis, "No association between preprocedural fasting and witnessed pulmonary aspiration: A systematic review and meta-analysis," garnered the most comments yet, primarily from biostatisticians and anesthesiologists. Most of the comments we received noted the serious nature of aspiration events, pointed out that the rarity of these events makes them very challenging to study, and highlighted that the American Society of Anesthesiologists (ASA) guidelines have already been in flux. To me, this supports the point I made in the podcast, that this meta-analysis opens the door for robust RCTs

Today, we are pleased to share one of the responses we received. Peter Nielson makes important points, and his essay is a good representation of what we heard.

Adam Cifu

As an anesthesiologist, I think about aspiration a lot. Listening to Drs. Cifu and Mandrola’s recent podcast, I sensed that they think the recent meta-analysis on pre-procedural fasting should change our practice. I disagree, and the devil is in the details of what we consider a “liberal” fasting strategy. I think listeners are likely to draw the incorrect conclusions about fasting guidelines because Cifu and Mandrola didn’t specify what exactly “liberal fasting” is and how it differs in this paper, or doesn’t, from existing ASA guidelines. It is also important to consider human factors before we liberalize fasting guidelines.

What “non-fasted” meant in the trials included in the meta-analysis is important to examine closely. Fasting can have various lay definitions, including abstaining from food, water, or both, but in medicine, we use the term NPO, which means "nothing by mouth." Many of the experimental groups had water or chewing gum as the “non-fasted” intervention. Some studies included tube feeds in the ICU in intubated patients. Other studies included patients who required emergency surgeries, who had not had time to fast.

For the studies making up most of the weight of Figure 2 (Bassa and Sharma below), the comparison was between intubated patients with a < 6-hour fast and a > 6-hour fast in one study, and a comparison between a 2-hour NPO status and NPO after midnight.

Importantly, these are not cases where a patient discloses that they had a slice of pizza before coming in, or half a breakfast when instructed not to have anything, and then randomized to have surgery or not. What this study shows me is that the ASA guidance of chewing gum or having clear liquids two hours prior to elective surgery is probably correct and evidence-based.

The ASA has actually allowed clear liquids for some time. They updated their guidelines in 2023 to address chewing gum and some other variations of clear liquids. This change may have prevented a lot of glycemic swings and potentially help patients from being in a catabolic state right before surgery.

These changes have also probably muddied the waters. These guidelines may not apply to someone taking a GLP1a or a patient with gastric motility problems. What about patients who don’t understand that “clear liquid, including black coffee” means clear liquid and black coffee, not a latte? Some patients aren’t given these more liberal instructions and get “NPO after midnight” and still eat something from midnight to 7 am the day of surgery.

Allowing clears in some situations is a step in the right direction, but it was a step already taken by many since the 2017 guidelines. It is easier to say “don’t eat or drink anything” than to provide nuanced guidance and then have to cancel a procedure when someone has a whey protein shake three hours before they arrive. Specific guidance on what constitutes a clear liquid or even “light meal” is important to tell patients explicitly, but I’ve seen even that backfire and lead to cancellation day of surgery.

It is worth noting that aspiration is a rare event. Its incidence is probably even lower in studies if the data originates from the EMR alone. The oft-quoted 1:1000 incidence translates to approximately two aspirations per year for each anesthesiologist. There may be many small aspiration events that go unnoticed and unreported, especially if they require coding in the EMR to be recognized.

As Dr. Cifu pointed out in the podcast, aspiration can be quite bad, and patients can end up quite ill. Aspiration of a large volume of solid contents can lead to severe morbidity and even death. Many of the studies in this meta-analysis included only about 100 patients in each arm. The largest had 500 patients in each arm. The experimental group was frequently still largely “fasted other than liquids”. The point that it would be hard to study aspiration because it is a rare event is a fair one. It would, at the very least, require very large studies.

I find the most exciting area of research is gastric point-of-care ultrasound. I love the phrase “why guess when you can look,” and it applies here. Admittedly, we are using a surrogate maker of stomach contents and correlating that with the risk of aspiration. The data I have seen usually references about 1.5cc/kg in the stomach as full, but there are some variations. I would argue that someone with a stomach full of solids is not optimized for elective surgery. Unfortunately, ultrasound is often performed in the moments before the decision to proceed, and guidelines should be in place to minimize day-of-surgery issues, both for convenience and safety.

The devil is in the details. Can most patients have clear liquids or chewing gum a few hours before surgery? Yes, they can, and this study supports the 2023 guideline update of the 2017 guidelines.

Personally, I don’t know that I would refer to this as a “liberal” fasting guideline, and I would be careful telling patients exactly what and when they can drink before the procedure. In cases of questionable gastric motility or when the gastric contents are uncertain, we utilize ultrasound to assess the risks associated with a full stomach. If we observe a full stomach, I would not proceed with suppressing someone’s airway reflexes unless absolutely necessary.

Peter Nielson, MD, is an anesthesiologist at the Lahey Clinic in Burlington, MA, who believes in using and teaching evidence-based medicine in all fields of medicine.