Title: Eating and drinking with an inflated tracheostomy cuff: a systematic review of the aspiration risk
Author: Diane Goff & Joanne Patterson
Journal: International Journal of Language & Communication Disorders
Year of Publication: 2018
Design Type: Systematic Review
Purpose: “This systematic review was undertaken in order to establish the available evidence regarding the aspiration risk when eating/drinking with an inflated tracheostomy cuff.”
Population: adults with tracheostomy
Inclusion criteria: English studies; full manuscripts; reporting aspiration rates in adult humans with inflated cuff tracheostomies
Exclusion criteria: non-English studies; expert opinion papers; conference abstracts; studies with primary outcome as cuff integrity
For the SLPs who have trach-on-the-brain or those less familiar but wanting to know more, this article attempts to shed a new light on what we knew, what we think we know, and what we now know about managing the “trachoestomy population” when it comes to swallowing function, aspiration, and inflated cuff.
“The ‘tracheostomy population’ by nature is a diverse group, and swallowing is a complex process that can be affected by any number of variables.” p.38
Forewarning: there are some vague statements early in the article about listening to vocal quality and other possible “clinical indicators” observed during clinical bedside exams for aspiration and use of Modified Blue Dye Tests that I had to try to ignore. I figured to let this slide in order to really get to the meat of the article. Another tiny thought to keep in mind throughout is that while our amazing colleagues across the pond are great, they do follow a different governing body (Royal College of Speech and Language Therapists, RCSLT) just to be aware so we’re not comparing Lays chips to french fries🙂 .
Luckily the article really does a great job of breaking things down as far as what we already know in regards to this topic, but here are a few tidbits to catch you up to speed (SLT=Speech Language Therapist):
“Currently, there is no agreement among SLTs on how to proceed with the assessment of swallow function for those patients unable to tolerate cuff deflation (Ward et al. 2007, McGowan et al. 2014)”
“Some clinicians wait until the patient can tolerate cuff deflation before assessing swallowing. This potentially leaves an alert patient who is keen to eat/drink, nil by mouth for longer which has financial implications due to prolonged enteral feeding and length of hospital stay and implications for the patient’s psychological well-being and quality of life (Sherlock et al. 2009, Foster 2010).” p. 31
Why could swallowing with an inflated cuff have negative consequences?
- reduced/absent airflow through upper airway decreases sensation (lack of cough reflex)
- “laryngeal tethering/anchoring” causing decreased laryngeal elevation for protection
- changes in swallowing physiology
- possible impingement onto shared tracheo-esophageal wall
While these are just to name a few (please talk to your colleagues/peers/researchers to learn more!), some are still up for debate. In my mind, this had always made sense to me by thinking of the cogs in a clock. There’s also the debate over inflated cuff as it relates to aspiration, which can sometimes feel like a lifelong battle between SLPs and some docs/medical staff when arguing about this😩😩 (PLEASE save mine and your sanity by doing a quick search to learn more about how an inflated cuff inherently does NOT protect from aspirated material!!!).
After that general crash course, here’s how the authors went about looking for adequate studies to help us fellow clinicians out…..
Using multiple, go-to search engines (e.g. CINAHL, Medline, etc.), they scoured through articles as far back as 1960s up to 2017. When trying to attempt a formal literature review in the past, I learned a lot more about the world of search terms and their effects. So the curious part of me couldn’t help but wonder, “Could other terms have been used besides ‘tracheostomy and deglutition disorders’ and free-text words ‘aspiration, dysphagia, swallowing and cuff’ such as “inflated cuff” or “oral diet” just to see if more results could have appeared? I mean, we definitely know from IDDSI how many of us use different terms for different things! I guess we’ll never know (or as I keep learning I’ll eventually find out the answer to my own question!😁)
While I didn’t necessarily have to do a quick Google search for “grey literature definition” because a) I had no clue what that meant (secretly hoping it was related to Grey’s Anatomy😅), and b) the article actually does give some good context clues, my results gave me some more info:
“The term grey literature refers to research that is either unpublished or has been published in non-commercial form. Examples of grey literature include: government reports, policy statements, and issues papers.” Google search; https://www.une.edu.au/library/support/eskills-plus/research-skills/grey-literature
Again, I also didn’t absolutely have to do a Google search when I came across “Downs and Black Checklist” because the authors so graciously included the reference and a quick description. But I couldn’t help my inner research nerd wanting to know more about different ways to evaluate different kinds of research (🤓plus I love checklists🤓).
“This is a 27-item tool that assesses five sub-scales: reporting, internal validity (bias), internal validity (confounding), external validity and power. It is a well- established tool developed using rigorous psychometric methods (Saunders et al. 2003) and tested for reliability and validity.” p. 32 (Original reference (Downs and Black 1998)
So, the next time you find an amazing article you believe to be the holy grail, doing a quick ✅check-off✅ using this resource can let you know if the article is actually all it says it is and truly powerful, or if it’s just pretending to be aposer article👥. . .
Similar to American Idol or Survivor, and just like any Systematic Review, there’s a process to wean out the weak to only end up with the best. The authors’ process went something like this:
After searching 511 + 10 additional grey literature and tossing out duplicates…….454……..37 (full paper review)………10 studies went for Final Review!!
How’s that for easy counting? The final 10 articles were a collection of different country locations (USA, UK, Japan) ranging from years 1973-2012 (doesn’t that seem so close but so long ago?!?).
How did they rate their findings?
See Oxford Centre for Evidence-Based Medicine’s (OCEBM) Levels of Evidence 1, which is similar to our more familiar ASHA’s Levels of Evidence. Basically, out of the 10 fully reviewed articles they ended up with, here’s how their results looked when assigning what+how many Levels of Evidence were found (in order from highest to lowest rated; 1=Randomized Controlled Trials/Systematic Review, 5=Expert Opinion, respectively):
NOTE: NO Randomized Control Studies were present. That may be surprising to some, and not even resulting in an eyebrow raise for others.
The authors also include the fact that due to the earlier dates of some of the published studies (1970s), some were not exposed to the more rigorous review process as far as quality goes now.
All 10 of the studies included a variety of populations, settings, and diagnoses. Try to picture 1 large pot full of beans. The only
commonality across all the different kinds is just the “bean” category (aka all patients in studies had tracheostomy). Now, there are a lot of regular beans out there (red, black, white, etc. etc.). You might have a different kind of bean for each of the following: blunt neuro/chest trauma, head/neck cancer, sepsis, respiratory failure. Next, you add in other varieties of beans such as green beans, string beans, etc. to represent ICU, inpatient rehabilitation, and radiology (aka the 3 different study settings). Now, you may have anywhere from 4 to 623 total beans in this pot (aka range of sample sizes across all studies). Finally, you realize that some beans taste better with certain spices, so there may be a variety of spices in your bean pot as well—these are the variety of methods that were used in all the final articles to assess aspiration, which included bronchoscopy, blue-dye testing, pulse oximetry, VFSS/MBSS, and FEES.
Now, I’m not even a fan of beans, but some of those ingredients just don’t sit well with me anyway. Meaning: the evidence for pulse oximetry and blue-dye testing are far and few between, not to mention the lack of consistency and quality of studies (check out Pulse Oximetry Systematic Review Blog Post), so the 2 studies using this measure were dubbed not reliable/valid due to the nature of the ‘lack-there-of’ argument.
Now that there’s a heck of a lot of stuff in your pot, next you may want to figure out what’s worth keeping and what’s not.
The authors looked at the different measures shown below from the Downs & Black checklist to evaluate the quality of their final studies:
Reporting– details clarifying anything from how patients were selected and remained in a study to outcomes, methods, and interventions
External validity– how well data and theories from one setting apply to another
Internal validity– how well an experiment is done, especially whether it avoids confounding°. The less chance for confounding in a study, the higher its internal validity is.
I promise you’ll want to know how they decided to show you which studies got a √ from the Downs & Black Checklist questions:
🙂 = yes (included)
🙁 = no (not included)
😐 = unable to detect
Yes, they literally used those symbols in the paper!! So more 🙂 = higher score = higher quality rating !!! This chart alone is definitely worth checking out, I seriously couldn’t make this up!! (also I’ve literally been saying for years why we can’t just write reports or articles using emojis, so this part seriously made my day😍🤪!).
What did they find?
Without going into too much detail, the quality of most-to-all studies just was not there.
“Of the 10 studies included in this systematic review, only one found a statistically significant difference between cuff conditions, Ding & Logemann 2005 found, however that the actual rate of aspiration was not significantly higher in the cuff-inflated group; silent aspiration was more prevalent.” p. 38
I wanted to take a shot at this interpretation—the cuff-inflated group was significant, “statistically speaking”
BUT not significant when compared with the cuff-deflated group (e.g. an apple is significantly round, but not when compared to an orange🍎🍊)
Silent aspiration was more significant within the cuff-inflated group. I do feel this helps us think a bit out of the box with this population and what we should be most concerned about a bit differently, so I’m excited to see any future research that answers more of these questions!
When reading over the rest of the results from the studies, it did feel more like Goldilocks and the 3 bears’ problems.
Some studies weren’t randomized (so you really can’t tell if there was bias like who got an inflated versus deflated cuff first, or if fatigue played a factor etc.). Some studies had poor detailed methods for what they did and/or how they did it (e.g. how patients were recruited, what/how much bolus trials were used, etc.). Others had no control groups or didn’t use any kind of statistical analysis for their results (…I didn’t think that was possible for research?🤨).
Why could any of this be important to consider when a clinician just wants to know what to generally do with this population?
- By not having a control group, you can’t compare anything except simply say “this happened”
- By not using any kind of statistical analysis, you can’t possibly confidently confirm “this happened because of x,y,z” etc.
- Without controlling for bias, who’s to say a researcher in a study may have been unconsciously more inclined to “see aspiration” on instrumental because he/she knew a cuff was inflated? (I figured we’d give a pass on blinding for this since it really would’ve been near impossible to blind the investigators for who’s inflated vs. deflated cuff, but hopefully you get my point)
- How ‘powerful’ a study is should also be a factor to think about:
“Statistical Power is the probability that a statistical test will detect differences when they truly exist. Think of Statistical Power as having the statistical “muscle” to be able to detect differences between the groups you are studying, or making sure you do not “miss” finding differences.” https://segmeasurement.com/content/what-statistical-power-and-why-it-important
This was a big defeat for one of the studies (David et al., 2002) because even though they found higher aspiration rates in cuff-inflated group, this did not reach significance. On the flip-side, their study also found cuff status to be a significant predictor of aspiration (p=0.032), BUT because the “power” of the study was weak, this overshadowed the ‘true’ significance they did find. The authors actually explain more limitations of the actual systematic review for any reader too glance over.
To wrap up, our UK friends brought it back to the main point very nicely while reminding of their governing body’s guidelines:
“A survey of UK SLTs found 33% would not assess swallowing function in this patient group (McGowan et al., 2014). As few as 7.5% said they would accept the referral with the remaining stating they ‘sometimes’ would accept a referral. It is not clear on what basis such decisions are being made.” p. 38
We are all SLPs/SLTs striving to be better every day for every patient in every situation. Which group percentage would you have answered before? Would the answer be different now based on this article’s review of the literature?
This article left me with a final question…if all other signs of risk for aspiration are low—meaning, patient is alert, intentionally communicating, more ambulatory, sufficient daily oral care, healthy dentition, consistently stable labs/chest X-ray, with possible inconsistent weaning status…..what then are the risks versus benefits for trying to advocate for instrumental assessment to evaluate swallow function in both conditions? How could you know for sure that the patient is absolutely 100% at risk for aspiration? Yes, there may absolutely still be risk, that’s typically why patients are in a medical facility in the first place. But do we, as medical professionals, feel we are able to give this kind of patient a shot?
How can you use this article?!?
I’m honestly so intrigued by what this article brought to light, and would love to know others’ thoughts!
Have you ever tried doing instrumental to see if a patient with inflated cuff can safely consume oral nutrition??
Do you have a strict policy that inflated cuff=NPO??
Or are you trying to change mindsets to treat each patient individually??
For now I don’t know the answer, because I don’t think there is a clear-cut one. But this article at the very least can start the conversation about different options and makes us better for thinking outside the box to provide holistic/pragmatic, individualized clinical care: ⬇️⬇️
“Referrals for cuff-inflated patients may need to be triaged on a case-by-case basis. The expertise of SLTs in managing dysphagia would indicate that their involvement at an early stage is beneficial, regardless of cuff status.” p. 39
- “Evidence regarding aspiration risk in adults with cuff-inflated tracheostomy is inconclusive.”
- “the decision to feed with an inflated cuff should be made on an individual patient basis”
- “All studies included in this review lacked consensus regarding methods of assessing aspiration. The suggestion of higher silent aspiration risk advocates the role for instrumental evaluation of swallowing, which will more reliably detect this (Hales et al. 2008, Ponfick et al. 2015)”
- “Currently, there is no agreement among SLTs on how to proceed with the assessment of swallow function for those patients unable to tolerate cuff deflation (Ward et al. 2007, McGowan et al. 2014)”
Goff, D. and Patterson, J. (2019), Eating and drinking with an inflated tracheostomy cuff: a systematic review of the aspiration risk. International Journal of Language & Communication Disorders, 54: 30-40. doi:10.1111/1460-6984.12430
***NOTE: This post was written long before COVID19, and thus information and conclusions cannot obviously be compared to many current situations. Please see recent posts regarding COVID19-related information and recommendations (Healthcare, Adults, COVID-19, and Acute Care in Hospitals: Recent Recommendations) and a variety of other resources/references****