Relief of pain, and in particular, relief of perioperative acute pain, is a core precept of anesthesiology, but safe and effective postoperative pain relief remains an elusive goal.1 Inadequate treatment of perioperative pain is associated with adverse outcomes, including thromboembolic and pulmonary complications, increased length of stay in intensive care unit (ICU) or hospital, more frequent hospital readmission, development of chronic pain, and worsened health-related quality of life. Opioids, still a cornerstone of acute pain management, are also associated with adverse outcomes, including, most seriously, sedation and respiratory depression resulting in death or severe brain damage.2 Use of multimodal analgesia is recommended to reduce opioid requirements, and there is some evidence that peripheral nerve blocks2 and neuraxial analgesia,3 for example, reduce the risk of adverse outcomes related to respiratory depression. The use of short-acting opioids, such as fentanyl, instead of long-acting agents, such as morphine or hydromorphone, for patient-controlled analgesia (PCA) has also been advocated as a way to reduce respiratory depression by synchronizing plasma levels of opioid with dynamic episodes of acute pain. In the current issue, Belcher et al4 examined the rate of postoperative hypoxemia in patients receiving short-acting versus long-acting opioids using single-center results from a 40,000-patient, multicenter, prospective cohort study (Vascular events In noncardiac Surgery patIents cOhort evaluatioN [VISION], NCT00512109)5 and found no evidence that short-acting opioids reduce postoperative hypoxemia.
In the study by Belcher et al,4 duration and severity of hypoxemia was evaluated in a subgroup of participants enrolled at a single center during the VISION trial who were monitored with continuous pulse oximetry and received exclusively either short-acting or long-acting opioids via PCA. Out of 1250 participants enrolled at the Cleveland Clinic, 191 were included in the analysis; 75 (39%) received only PCA fentanyl, and 116 (61%) received only PCA morphine or hydromorphone for postoperative analgesia. Continuous pulse oximetry data were not available to caregivers, patients, or families. Multivariable analysis was used to adjust for 7 prespecified expected confounders (type and duration of surgery, American Society of Anesthesiologists Physical Status, home oxygen use, pre-existing pulmonary disease, previous congestive heart failure, and intraoperative opioid dose), and 11 variables were identified as imbalanced between the groups (including age, history of aortic stenosis, diabetes, sleep apnea, hypertension, open surgery, and total dose of neuromuscular blocking agent intraoperatively). There was no difference between groups in the integrated area under the curve for duration of arterial oxyhemoglobin saturation by pulse oximetry (SpO2) during the first 2 postoperative days, regardless of cut point (SpO2 < 95%, SpO2 < 90%, or SpO2 < 85%). Overall, about three-quarters of participants in each group had an SpO2 of < 95% for at least 10 minutes per hour, and just more than half had an SpO2 of < 95% for 20 minutes per hour. About 20% of participants in each group had an SpO2 of < 90% for at least 10 minutes per hour. The cut points selected were somewhat arbitrary. Few studies have examined the relationship between perioperative SpO2 and outcome, although an SpO2 of < 95% has been associated with an increased risk of surgical site infection,6 and postoperative hypoxemia has been associated with myocardial ischemia7 and cognitive dysfunction.8
The study by Belcher et al4 provides a valuable contribution to the conversation about reducing opioid-related morbidity while providing effective analgesia, but it is not definitive. The final cohort analyzed (191 participants) was a minor subset of the original cohort. Because the study was observational, the choice of analgesic was not randomized. The data suggest that providers already had informal criteria in place for opioid selection: participants who received short-acting opioids tended to be less healthy, and participants who received long-acting opioids tended to have longer, more painful surgical procedures. Although this was accounted for in the data analysis, selection bias in assigning opioids could have skewed the results. Major confounders not accounted for in data analysis were the timing and duration of the supplemental oxygen administration, which were not recorded during data collection. Supplemental oxygen would be expected to reduce hypoxemia in patients receiving IV opioids.9 Given the existing perception that long-acting opioids might be associated with more hypoxemia, participants who received long-acting opioids may also have been more likely to receive supplemental oxygen.
The take-home message is that regardless of whether long-acting or short-acting opioids are prescribed, postoperative hypoxemia is common, and severe, prolonged hypoxemia may be undetected and untreated.9,10 If short-acting opioids are not the answer, then what is? Five areas show promise for improving postoperative management and reducing the incidence of respiratory depression: (1) multimodal analgesia and reduction of opioid dose2,3,11; (2) improved monitoring and feedback,1 including both continuous pulse oximetry and measures of ventilation such as continuous capnography; (3) better identification of risk factors for postoperative hypoxemia;1,12 (4) more effective education of physicians, nurses, health care assistants, patients, and families;2,13 and (5) provision of continuous supplemental oxygen to all patients receiving opioids.9,14–16
Multimodal analgesia appears to provide some protection by reducing the opioid dose. Replacing opioids with safer and equally effective analgesics would be ideal, but it has proven to be an elusive goal.11
Routine, continuous pulse oximetry in all patients receiving postoperative opioids has long been recommended, but adoption has been slow and incomplete.17 False alarms are common, and patients could remove the monitor. Although alarms may alert the patient and remind them to breathe, they are also annoying and may interrupt sleep. Alarms are not audible at the nursing station, so a staffed central monitoring station is required if unit staff are expected to respond. Continuous bedside capnography is useful for identifying hypoventilation, although it does not provide a direct measure of oxygenation. Limitations are similar to those of pulse oximetry. Currently available devices tend to be bulky and inconvenient. Many devices that are compact, easy to use, and provide feedback when hypoventilation is detected are in development.
Risk factors for postoperative hypoxemia identified in the current study4 align with those commonly reported in the literature1 and can help to identify patients for whom additional monitoring is indicated. Genetics clearly contribute to variability in opioid requirements and side effects; in the future, genomic analysis could be used to optimize postoperative analgesic selection and dosing.12
In a recent analysis of the Anesthesia Closed Claims Project database,2 patients were described as somnolent before the event in 60% of 92 episodes of respiratory depression that led to a malpractice claim, including 9 patients assessed within 15 minutes of the event. In a previous analysis of the VISION data set used for the current study,10 floor staff did not document 90% of hypoxemic episodes in which SpO2 was <90% for at least 1 hour. This suggests a role for better education of health care providers to improve recognition of respiratory depression and to implement measures to correct and prevent hypoxemia.
Pulse oximetry measures arterial oxygenation, and it is not a sensitive monitor of hypoventilation when supplemental oxygen is given; patients breathing as little as 25% to 30% supplemental oxygen usually maintain arterial oxygen saturation even in the face of severe hypoventilation.14–16 Opioids reliably induce hypoventilation, and there are few adverse effects of mildly elevated arterial carbon dioxide.18 It is reasonable to focus on maintaining SpO2 in the face of hypoventilation rather than to focus primarily on preventing hypoventilation. Therefore, a solution to the problem of opioid-induced postoperative respiratory depression is to provide continuous supplemental oxygen for all patients receiving opioids in the hospital9 or to provide supplemental oxygen for all patients who do not maintain an SpO2 of > 94% of breathing room air and are unstimulated.19 Supplemental oxygen is inexpensive, relatively easy to apply, and there is evidence of its use that it improves patient outcomes.6–8 One barrier to the implementation of routine postoperative supplemental oxygen is that patients find supplemental oxygen uncomfortable and difficult to keep in place, and it may reduce mobility. Another barrier is the common practice of using an oxygen requirement as a measure of progress toward discharge. Patient and provider education could overcome this barrier by emphasizing the use of supplemental oxygen as a safety measure that helps the patient heal.
Opioids remain a cornerstone of postoperative analgesia. Postoperative respiratory depression is a common, potentially serious side effect of opioid administration that frequently results in hypoxemia. Numerous long-term solutions should be pursued, including approaches that reduce the need for opioids, development of improved monitoring, and development of potent analgesic agents that do not cause respiratory depression. In the meantime, vigilance for signs of impending respiratory arrest and routine use of pulse oximetry, respiratory monitoring, and postoperative oxygen supplementation should be considered in all hospitalized postoperative patients who are receiving opioids.
Name: Harriet W. Hopf, MD.
Contribution: This author wrote the manuscript.
This manuscript was handled by: Richard C. Prielipp, MD.
1. Practice guidelines for acute pain management in the perioperative setting.Anesthesiology2012116248–273
2. Lee LA, Caplan RA, Stephens LS, et al.Postoperative opioid-induced respiratory depression: a closed claims analysis.Anesthesiology2015122659–665
3. Flisberg P, Rudin A, Linner R, Lundberg CJPain relief and safety after major surgery. A prospective study of epidural and intravenous analgesia in 2696 patients.Acta Anaesthesiol Scand200347457–465
4. Belcher AW, Khanna AK, Leung S, et al.Long-acting patient-controlled opioids are not associated with more postoperative hypoxemia than short-acting patient-controlled opioids after noncardiac surgery: a cohort analysis.Anesth Analg2016
5. The Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Study InvestigatorsAssociation between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery.JAMA20123072295–2304
6. Govinda R, Kasuya Y, Bala E, et al.Early postoperative subcutaneous tissue oxygen predicts surgical site infection.Anesth Analg2010111946–952
7. Gill NP, Wright B, Reilly CSRelationship between hypoxaemic and cardiac ischaemic events in the perioperative period.Br J Anaesth199268471–473
8. Rosenberg J, Kehlet HPostoperative mental confusion—association with postoperative hypoxemia.Surgery199311476–81
9. Stone JG, Cozine KA, Wald ANocturnal oxygenation during patient-controlled analgesia.Anesth Analg199989104–110
10. Sun Z, Sessler DI, Dalton JE, et al.Postoperative hypoxemia is common and persistent: a prospective blinded observational study.Anesth Analg2015121709–715
11. Corbett AD, Henderson G, McKnight AT, Paterson SJ75 years of opioid research: the exciting but vain quest for the Holy Grail.Br J Pharmacol2006147suppl 1S153–S162
12. De Gregori M, Diatchenko L, Ingelmo PM, et al.Human genetic variability contributes to postoperative morphine consumption.J Pain201617628–636
13. Lynn LA, Curry JPPatterns of unexpected in-hospital deaths: a root cause analysis.Patient Saf Surg201153
14. Fu ES, Downs JB, Schweiger JW, Miguel RV, Smith RASupplemental oxygen impairs detection of hypoventilation by pulse oximetry.Chest20041261552–1558
15. Niesters M, Mahajan RP, Aarts L, Dahan AHigh-inspired oxygen concentration further impairs opioid-induced respiratory depression.Br J Anaesth2013110837–841
16. Rozario L, Sloper D, Sheridan MJSupplemental oxygen during moderate sedation and the occurrence of clinically significant desaturation during endoscopic procedures.Gastroenterol Nurs200831281–285
18. Contreras M, Masterson C, Laffey JGPermissive hypercapnia: what to remember.Curr Opin Anaesthesiol20152826–37
19. O’Driscoll BR, Howard LS, Davison AGBritish Thoracic SocietyEmergency oxygen use in adult patients: concise guidance.Clin Med201111372–375