Patient Safety Tip of the Week

December 6, 2011

Why You Need to Beware of Oxygen Therapy



Our regular readers are probably tired of hearing us harp on the dangers of using oxygen in patients who are on PCA pumps or receiving intravenous opiates. But we by chance stumbled onto an old paper that illustrates the hazard much better than we could ever do.


Smyth and Egan (Smyth 1998) did a case report of a “near miss” in a patient on a PCA pump who was also receiving oxygen. The patient was in a neurosurgical critical care unit with acute spinal cord impingement and had been placed in tongs for cervical traction. He was receiving morphine via PCA pump. The original PCA order had been written by the Pain Management Service and was for 1 mg of morphine with a 10-minute lockout interval, later increased to 1.5 mg of morphine. Later, another physician ordered two 5 mg doses of diazepam for anxiety (two hours apart). After the second diazepam dose there was a brief period where oxygen saturation fell to the 85-88% range. His supplemental oxygen was increased from 3 L/min via nasal cannula to 10 L/min via face mask. About an hour later he was found unresponsive and apneic but oxygen saturations were in the 92-95% range. But arterial blood gases showed a pH 7.08 and pCO2 102 and pO2 90. He was treated with IV naloxone and bag-mask ventilation and after 10 minutes became alert with spontaneous respiration.


Obviously the pulse oximetry was not helpful in this case since his oxygen saturation did not fall to levels where an alarm would be triggered. Similarly, the other alarm systems in the critical care unit (EKG and automated BP monitors) were not helpful in alerting nursing that there was a problem with this patient.


The authors discuss the physiology of how oxygenation is maintained in someone who is apneic, hence the term “apneic oxygenation”. This comes as no surprise to those of us who are neurologists since we routinely use tracheal insufflation of oxygen during apnea testing as part of braindeath determinations. Very seldom do patients develop significant oxygen desaturation during braindeath apnea testing.


But the key lesson learned here regards the false sense of security one gets when using continuous pulse oximetry monitoring. Obviously it was of no benefit here. The other key lesson learned regards the additive effects of a benzodiazepine and morphine in producing respiratory depression.


But there are a few other unanswered questions. We were not told why the patient was put on oxygen in the first place (though he had a 70-pack year smoking history). Secondly, we don’t know what sort of evaluaton took place when the patient initially had the brief episode of oxygen desaturation that led to increasing his supplemental oxygen. Unfortunately, a common response to oxygen desaturation is simply increasing the amount of supplemental oxygen without doing an assessment of why the desaturation occurred. In their excellent paper on patterns of unexpected death in hospitals Lynn and Curry (Lynn 2011) note that in pattern #2 (CO2 narcosis) it is not uncommon to see overlap with pattern #3 (a sleep apnea pattern). They also note that it is not at all uncommon for nurses to respond to frequent brief episodes of oxygen desaturation in such patients and consider them “false alarms”. Such episodes might, in fact, be sleep apneic episodes and the stimulation by nurses wakes the patient so their breathing and O2 saturations are restored and the episodes are thus considered “false”.


We’ve noted several times the excellent article by Lynn and Curry  (Lynn 2011) on alarms and their failure to identify deteriorating patients early, highlighted in our February 22, 2011 Patient Safety Tip of the Week “Rethinking Alarms”. That article and the APSF initiative on monitoring the postoperative patient have really challenged the current methods of monitoring we use and pointed out the weaknesses of threshold-based alarms.


The APSF (Anesthesia Patient Safety Foundation) summary of their “Conclusions and Recommendations from June 08, 2011 Conference on Electronic Monitoring Strategies to Detect Drug-Induced Postoperative Respiratory Depression” made a cogent argument for continuous physiological monitoring of all patients receiving post-op opiates, including continuous pulse oximetry (and capnography for all patients receiving supplemental oxygen). And though they note the importance of identifying patients with conditions that place them at higher risk of respiratory depression, they note that selective monitoring of these would miss respiratory depression in patients without these risk factors. They again make the case that threshold-based alarm systems remain problematic and stress the need for systems that allow for earlier recognition of respiratory depression.


Obviously the current case highlights the need for new monitoring and intervention tools such as the apnea prevention device (Zornow 2011) mentioned in our September 6, 2011 Patient Safety Tip of the Week “More Tips on PCA Safety”.


Sometimes we do things in medicine that superficially sound good – like putting post-op patients on oxygen – only to see unintended consequences that are patient safety hazards. This is one of the reasons we like checklists or similar formats when using PCA pumps or otherwise treating post-op patients with opiates. These can force the ordering clinician to think twice before ordering oxygen in such patients and to consider whether appropriate monitoring protocols are in place.





Other Patient Safety Tips of the Week pertaining to opioid-induced respiratory depression and PCA safety:




Prior columns pertaining to oxygen safety:


·        April 8, 2008 Patient Safety Tip of the Week “Oxygen as a Medication

·        June 10, 2008 Patient Safety Tip of the Week “Monitoring the Postoperative COPD Patient

·        January 27, 2009 Patient Safety Tip of the Week “Oxygen Therapy: Everything You Wanted to Know and More!

·        April 2009 What’s New in the Patient Safety World column “Nursing Companion to the BTS Oxygen Therapy Guidelines”.






Smyth E, Egan TD. Apneic Oxygenation Associated with Patient-Controlled Analgesia. Journal of Clinical Anesthesia 1998; 10: 499-501



Lynn LA, Curry JP. Patterns of unexpected in-hospital deaths: a root cause analysis. Patient Safety in Surgery 2011, 5:3 (11 February 2011)



APSF (Anesthesia Patient Safety Foundation). Conclusions and Recommendations from June 08, 2011 Conference on Electronic Monitoring Strategies to Detect Drug-Induced Postoperative Respiratory Depression. 2011



Zornow MH. Clinical Testing of the Apnea Prevention Device: Proof of Concept Data. Anesth Analg 2011; 112:;582-586
















Tip of the Week Archive


What’s New in the Patient Safety World Archive