Title: Utility of Pulse Oximetry to Detect Aspiration: An Evidence-Based Systematic Review
Authors: Britton, D., Roeske, A., Ennis, S.K. et al.
Year of Publication: 2018
Design Type: Systematic Review
Purpose: “The purpose of this study is to systematically review evidence on the use of pulse oximetry in individuals with dysphagia to detect a decrease in SPO2 indicating aspiration during swallowing, toward the goal of further informing clinical practice in dysphagia assessment”
Population: adults with dysphagia
Inclusion criteria: studies focusing on population of adults with dysphagia; only studies that simultaneously used gold standard instrumental evaluations (i.e. VFSS/FEES) in conjunction with pulse oximetry desaturation to detect aspiration
Exclusion criteria: pediatric population; non-English studies; conference abstracts; studies with use of pulse oximetry mixed with other methods (due to difficulty to piece apart what was effective)
I remember first hearing of pulse oximetry during the start of my clinical fellowship at a skilled nursing/sub-acute facility, as it was recommended by the previous SLP to purchase one, especially with the trach/vent patients. I then remember my clinical instructor drilling into my head to “assess the patient, what are they physically doing/looking like?” versus relying on any gadgets or equipment. Were they breathing? Did they look like they were in distress?
Fast forward to years later and yes, you will find a pulse oximeter in my SLP bag or scrub pocket, but mainly to monitor overall cardio-pulmonary status. I’ll also absolutely look at this reading on the monitor along with other vitals like respiratory rate (RR), heart rate (HR), blood pressure (BP), etc. when in ICU.
But do I rely on it? No. Do I use ANY instrument/equipment/tool/number/measure as a sole indicator of aspiration, dysphagia, or any speech language pathology-related diagnosis? No. We as a field are better than that (or at least becoming better).
So why do we use it?
- gives you a pretty “quantitative” number to look at
Why have we wanted it to work so badly?
Because as human beings, we tend to go towards the path of least resistance or the easiest way to do something. So if we are given a magical ball that lets us “see into the future” or tells us “there’s aspiration!” and it’s cheap, heck-to-the-yeah we’re gonna get our hands on it.
Instead of taking this method at surface value and holding our breaths that it helps (pun intended), we need to dive a little deeper….
“Pulse oximetry is a commonly used noninvasive means of measuring peripheral capillary oxygen saturation (SpO2), i.e., the percentage of oxygenated hemoglobin in individuals at risk for hypoxemia. This is accomplished by measuring the ratio of light absorption of oxygenated (oxyhemoglobin) and deoxygenated (deoxyhemoglobin) blood. The underlying rationale posited for use of pulse oximetry to detect aspiration is that aspiration may lead to bronchoconstriction and/or airway obstruction leading to ventilation–perfusion mismatch, which in turn results in a drop in oxygen saturation.” p. 282
When I first read and heard that this method essentially used light to tell me if my patient was aspirating material into their trachea and lungs, an eyebrow was raised no doubt, because you don’t frequently hear “light” and “aspiration” in the same sentence. But alas, my younger self proceeded to place my pulse oximeter on any and all fingers of my patient to help give me any piece of the swallowing puzzle I was so desperate to put together.
I think of systematic reviews like the daunting task of rummaging through your closet in search of a specific item, then throwing anything and everything around until you find JUST THE RIGHT ONE you were looking for. Basically, there is a LOT of internal processing, analyzing, and scrutinizing going on to ensure selection of the best/right item(s).
A quick reference about ASHA’s and Autralia’s Levels of Evidence criteria in comparison because the latter is what was used:
This systematic review used multiple, widely recognized search engines (e.g. PubMed, CINAHL Plus), search terms from anything/everything under the sun (okay, maaybe they could’ve included “cookie test” in the search since I still see it once in a blue moon, but other than that they’ve got it covered), and had investigators involved by physically going through their own “literature libraries” in case anything they thought was relevant may be missed. The overall agreement between reviewers was greater than 80% and blinded (not 100%, but not terrible), which eventually resolved with third-party blinded reviewers and discussion consensus.
After a total of 294 articles were found excluding duplicates, they then narrowed their search to a total of 10 articles (one for each finger!) (see full article for further details into the review criteria). This means that these 10 articles are the best of the best they could find. They did all the dirty work by critically evaluating ALL articles that would pop up in a search, then selecting only those that met the rigorous criteria above. This essentially saves YOU time by not having you review something that is either a) poor quality, or b) unrelated to what you’re after. These 10 articles are like the valedictorians of the school, the graduates of a tough military academy, or Forbe’s Top 10 Companies in the country. All go through a robust, systematic grueling process where the goal is to try to rip a study apart, so props to the Top 10!
Here we go………
Ultimately, the findings equated that while all the 10 studies looked at using pulse oximetry for drops in SPO2 to identify aspiration simultaneously with gold standard instrumental evaluation (i.e. VFSS/FEES), and were rated as Level III-2 based off Australian criteria for diagnostic accuracy, there was a LOT of inconclusive and inconsistent findings, interpretations, and criteria within each study. Not only did the review conclude that the majority of the studies themselves were inadequate (i.e. no comparison groups, variable/non-specific exclusion criteria, heterogeneous populations with multiple concurrent medical diagnoses), but the mere parameters inherent in the use of pulse oximetry paralleled this trend as well:
After a total of 294 articles were found excluding duplicates, they then narrowed their search to a total of 10 articles (I recommend the article for further details into the review criteria)! This means that these 10 articles are the best of the best that they could find. They did all the dirty work by critically evaluating ALL articles that would pop up in a search, then selecting only those that met the rigorous criteria above. This essentially saves YOU time by not having to review something that is either a) poor quality, or b) not related to what you’re after. These 10 articles are like the valedictorians of the school, the graduates of a tough military academy, or Forbe’s Top 10 Companies in the Country. All go through a robust, systematic grueling process where the goal is to try to rip a study apart, so props to the Top 10!!
(please see article reference for further detai
Specific issues within studies included:
Pulse Oximeter parameters:
- definition of what “Desaturation” even means (How much of x = desaturation?)
Results across studies revealed variability in criteria for what percentage drop in SpO2 was used (2-4% range),
“the majority specified a > 2% drop in O2 saturation during eating as ‘desaturation’. The stated rationale for this criterion is unclear and may have continued over time through precedent. The problem with this criterion is that the standard of error of measurement for pulse oximeters is generally about 2%.”p. 288
Margin of error (aka the amount that is allowed in case something is miscalculated) in most pulse ox’s is 2%, which is what some studies were using as their criteria to base aspiration off (basically saying this gadget has “x” amount of error, and then stating that EXACT error is what to base your conclusions off!)
- lack of or unclear efforts to prevent equipment malfunction/interpretation errors (aka factors which may contribute to SPO2 changes/errors such as movement, nail polish, cold hands/body temperature, anemia, carbon dioxide exposure, reduced peripheral blood flow)
- lack of blinding observations (which means investigators may have already known a patient’s diagnosis or if a patient had aspirated on instrumental examination, etc.)
- lack of control for viscosity/texture
- lack of comparison group
“bias and precision of SPO2 measures tend to worsen when SaO2 is lower than 90%..It is therefore plausible that an individual’s underlying respiratory function could impact the accuracy of pulse oximetry in detecting aspiration”p. 289
Instrumental Swallow Study Parameters:
- how investigators defined “aspirators” vs “non-aspirators” was either unclear or variable (basically how they clinically concluded what meant “aspiration” and “penetration” on instrumental)
- lack of standard protocols for instrumental evaluations (yes, we know this is already a ‘thing’)
Across the 10 articles reviewed: A sensitivity (meaning the pulse oximeter accurately detected patients who did aspirate) ranged from 10-87%. The specificity (meaning the pulse oximeter correctly pushed aside all patients who had NOT aspirated) ranged from 39-100%. Now, would you want to use something on a patient that could tell you they are aspirating, maybe 10% of the time? Or gamble to know they are definitely NOT aspirating maaaaybe 39% of the time?
“7/10 of the studies failed to demonstrate the association between aspiration and O2 desaturation”
“2/10 of the studies concluded that there was NO significant relationship between events of aspiration and subsequent O2 desaturation”
“Although the research findings are mixed, the majority of the studies examined in this systematic review do not support the use of pulse oximetry for purposes of detecting aspiration. In addition, the strength of the evidence is weakened in many of these studies by study design flaws and potential for bias, such as lack of a comparison group, blinding and inter-rater reliability, as well as inadequate sample size. Therefore, the evidence for use of pulse oximetry as a means to detect prandial aspiration is inadequate and not convincing.” p.286
I do want to mention that one study (Collins & Bakheit) showed a positive relationship between prandial desaturation and aspiration when given a “much larger bolus.” This should be taken with a grain of salt, given a) what we now know about larger bolus sizes plus the variability within dysphagic population and b) again, what we know about the inherent limitations for pulse oximetry and the quality of research for this topic.
Now, NOT to say Do Not Use This At Any Time For Anything (this is NOT what I’m saying!).
Pulse oximetry can be a very useful and helpful tool in assessing a patient’s overall medical status in relation to heart rate and general oxygenation….heck PTs and OTs use them to assess just that (I’ve been guilty of borrowing theirs when in a pickle)! So yes, by all means, have this in your SLP toolbox, in particular for those who you feel the need to monitor overall status or endurance. But NOT to solely evaluate your patient’s swallowing ability, let alone if an aspirational episode just happened on your nails-painted-cold-hand-smoker-anemic patient.
So instead of waiting for a reading from your pulse ox to see if they just aspirated (which one study showed took up to 10 minutes!), go ahead and use that time to schedule that instrumental evaluation with your dear peer for VFSS/FEES 😉
How can you use this article?!?
What are your experiences with pulse oximetry and your patients in different settings? What other ways do you find this useful rather than merely “aspiration detectors?” Or have you tossed something else aside that you once relied on in your “uninformed years?” (no shame, we’re all guilty of something)
- pulse oximetry is NOT a reliable or effective way of seeing if a dysphagic patient has aspirated, regardless of viscosity (particularly when simultaneously compared to instrumental evaluations)
- the majority of studies focusing on using pulse oximetry to detect aspiration are of very poor quality in not only design and methods, but also when put up against the Levels of Evidence
- pulse oximetry use for dysphagia/aspiration purpose is unreliable and variable across populations
- definitions for what ‘desaturation’ means for use with swallowing is inconsistent and sometimes equal to an inherent margin of error versus actual physiological response
- some physiologic factors may affect O2 desaturation such as larger boluses; however, O2 desaturation “associated with eating/swallowing-without aspiration-was observed in some of the reviewed studies”
- patients with lower baseline SPO2 from underlying respiratory dysfunction (or increased RR) could affect the accuracy of pulse oximetry reading for aspiration
Britton, D., Roeske, A., Ennis, S.K. et al. Dysphagia (2018) 33: 282. https://doi.org/10.1007/s00455-017-9868-1