This is the second article in a series exploring the impact of pulse oximetry alarm thresholds in hospitalized patients. In the first article, “Improving the Safety of Post-Surgical Care,” I introduced the concept that, although the current approach to physiologic threshold monitoring (triggering an alarm when oxygen saturation falls below 90%) works great in the OR, it is unreliable on post-surgical floors.
In this post, I will explore in more depth why the threshold for triggering a pulse oximetry alarm varies depending on the site of care.
Pulse oximetry in highly controlled settings
Pulse oximetry is an outstanding technology that is invaluable in surgical suites, post anesthesia care units, and critical care units where patients have a high risk for sudden, life-threatening airway loss. People cared for in those settings are usually heavily sedated or fully anesthetized and they are often on mechanical ventilators. A threshold breach of 90% oxygen saturation that triggers an alarm has proven useful in those settings, particularly because professionals trained in advanced airway management are always nearby to respond.
The situation on post surgical floors is quite different. Here the risks are most likely to be related to insidious onset of sepsis, aspiration, congestive heart failure, pulmonary embolus, and two different types of opioid-associated respiratory depression.
Another common risk on post-surgical floors is obstructive sleep apnea (OSA). It is estimated that 80 million Americans have OSA. In addition, a study has shown that immediately following surgery, a significant number of patients initially testing negative for OSA (81%) convert to having moderate to severe OSA while recovering in the hospital. This conversion is not entirely understood, but has been associated with opioid pain management.
Regardless, this high OSA incidence assures that pulse oximetry alarms set at 90% thresholds will sound far too frequently on post-surgical floors because of so many self-correcting, innocent sleep apneas that cause brief, repetitive desaturations breaching 90% while patients sleep.
A Dartmouth study published in 2010 showed that post-surgical patients average 6% of their sleep time desaturating below 90% resulting in significant alarm fatigue manifested by nursing response delays and their outright ignoring the alarms altogether.
The researchers cleverly adjusted their post-surgical oximetry alarm thresholds to avoid these false alarms. Nurses were notified when the modified alarm limits were breached via a wireless pager. Consequently, Dartmouth became the only institution with published results that that has managed to universally monitor their post-surgical patients. They have significant outcome improvements to show for it, including no opioid associated severe injuries or death since deploying their threshold adjustment in 2007.
Why isn’t everyone doing it?
Regardless of these outstanding results, the majority of the country remains oblivious to Dartmouth’s work. Clinicians continue to rely on 90% thresholds, even arguing that it is a Standard of Care because it is used the most, regardless of how incorrectly and incapably.
We seem amnestic to the purpose of all thresholds: That they serve as useful tools that can help us prevent patient harm. We forget that the only oximetry threshold ever proven to work on post surgical floors has been Dartmouth’s in 2010. We are facing a conundrum based upon absurdity that has literally been killing patients because of obdurance and ignorance.
As a result, opioid-related death and severe anoxic brain injury has continued unabated on post-surgical floors contributing to the estimated 20,000 opioid-related hospital adverse events occurring each year. All of this amidst clear, though ignored warnings, from the American Society of Anesthesiology and the Anesthesia Patient Safety Foundation.
How can the conundrum be fixed?
To fix the problem, we must first understand how the six emergent conditions (sepsis, aspiration, congestive heart failure, pulmonary embolus, and two types of opioid-associated respiratory depression) most likely encountered on post-surgical floors affect normal respiratory physiologic patterns. Only then can we “know the enemy” and detect lethal threats early enough that they can be managed without morbidity.
The good news is that even though there are six conditions, there are just three distinct patterns of unexpected, yet lethal, rapidly evolving conditions to consider. I will cover them in detail in the next post.
- Although a threshold of 90% oxygen saturation works well in the controlled environment of ORs, PACUs, and CCUs, it is too high for post-surgical floors and leads to false alarms and alarm fatigue.
- Adjusting alarm thresholds to avoid false alarms and then universally monitoring all triggered alarms improves patient safety.
- Only six conditions and three patterns characterize the most common types of respiratory dysfunction found on post-surgical floors. They must be understood to improve outcomes.