Capnography Monitoring Enhances Safety of Postoperative Patient-Controlled Analgesia

June 2008, Vol 1, No 5 - Clinical
Thomas McCarter, MD, FACP
Zakir Shaik, MD, MHA
Keith Scarfo, DO, MS
Laura J. Thompson, RT(R), MBA
Download PDF

Opioid-related respiratory depression can have tragic consequences. Failure to detect a patient's declining respiratory status before progression to respiratory depression can lead to unwarranted outcomes and possible need for critical care.1 Concerns about respiratory depression can also prevent clinicians from adequately treating postoperative pain.2 Patient-controlled analgesia (PCA) is widely used for postoperative opioid administration, because of its considerable potential to improve pain management by enabling patient control of dosing frequency.

PCA, however, is associated with significant hazards, 1,3-9 and the Anesthesia Patient Safety Foundation (APSF) has noted a significant, underappreciated risk of serious injury from PCA in the postoperative period. 10 The APSF has urged healthcare professionals to consider the potential safety value of continuous monitoring of oxygenation and ventilation in patients receiving PCA or neuraxial opioids during the postoperative period. According to the APSF, "It is not possible to reliably identify those patients at higher risk, and, therefore, all patients receiving parenteral opioids should be monitored."11

Postoperative patients typically are monitored by pulse oximetry, but a growing body of evidence supports the use of capnography for earlier and more reliable warnings of potential respiratory depression compared with pulse oximetry.12-16 In one study, pulse oximetry identified only 33% of patients with respiratory distress, whereas capnography identified 100% of patients.17

Capnography has also identified a much greater rate of respiratory depression by bradypnea than was previously thought. Compared with the 1% to 2% incidence reported by analysis of published data,18,19 a research study using capnography showed a 41% incidence of respiratory depression by bradypnea (defined as respiratory rate less than 10 breaths per minute [bpm] for 3 minutes or longer).20

Until recently, continuous capnography required the patient to be intubated and was mostly limited to use in critical care.12 Innovative sidestream capnography can reliably be used now in intubated and nonintubated patients, adult and pediatric, as well as patients receiving oxygen.21 In October 2006, Main Line Health upgraded the "smart pump" technology at 3 of our acute care hospitals to include computerized PCA infusion with dosing-error–reduction software and sidestream capnography—end-tidal carbon dioxide (EtCO2)—on a single platform. This article reviews a case series of patients hospitalized at our facilities between October 2006 and March 2007, reporting the results of capnographic monitoring in patients receiving PCA therapy in the postoperative setting.

Implementing Capnographic Monitoring of PCA

Main Line Health is an integrated healthcare system in suburban Philadelphia, PA, that includes 4 acute care hospitals licensed for 1047 beds. Recognizing opioids' potential for harm, our organization has implemented a variety of measures to improve the safety of PCA, including policies and procedures to be used for the selection of candidates for PCA, prescription order sets to minimize prescription-related errors, staff training and supervision to prevent programming-related errors, patient education to avoid "PCA by proxy," nurse-controlled analgesia in case the patient is not able to administer the drug without assistance, and close monitoring to detect early signs of respiratory depression.

A multidisciplinary team was responsible for upgrading the existing smart pump system (Alaris System with Guardrails suite of safety software; Cardinal Health, San Diego, CA; with Oridion Microstream capnography technology) by adding capnographic monitoring modules. The capnography modules were to be included on the same technology platform as PCA pumps. The team included members of the medical staff, nursing, pharmacy, respiratory therapy, biomedical engineering, information systems, and quality. Team members were responsible for reviewing information about the new technology, planning its implementation, and sharing decisions with their peers at all acute care hospitals, except for the hospital that had only recently become part of Main Line Health.

The team also developed educational materials to train staff on EtCO2 technology. After implementation, the team met weekly to review any unresolved issues and track progress. Respiratory rate and EtCO2 minimum and maximum levels were determined after speaking with representatives from other healthcare systems that had implemented this technology.


Capnography modules provided real-time data on ventilatory status of postoperative patients receiving PCA therapy by measuring respiratory rate, apneic events, and concentrations of EtCO2. A nasal sampling cannulalike device was used to measure carbon dioxide (CO2) in exhaled breaths (Figure). Since any amount of CO2 in an exhaled breath was indicative of gas exchange, respiratory rate data were assumed to be representative of the patient's respiratory rate, as long as the sampling cannula was properly positioned.20 An alert was generated and a PCA "pause" protocol could automatically halt the opioid infusion if the respiratory rate was below the pre-established parameters (Table 1).

Table 1
Table 1

Patients using PCA were also monitored for vital signs, nausea, sedation level, pain scale, PCA dosing, and oxygen saturation as measured by pulse oximetry (SpO2).

Every PCA module was physically attached to a capnography module so that nursing could more easily adhere to the policy of using capnography to monitor every patient placed on PCA. This eliminated the need for nursing or central supply to search for a missing PCA or capnography module if the two were not together. The only exceptions to this practice were for patients receiving PCA while in active labor and for patients on palliative or end-of-life care.

A respiratory event was defined as occurring when the capnography monitor alerted the nurse (pulse oximetry did not alarm in any of the cases) to a change in respiration (apneic period or low respiratory rate) or to EtCO2 levels outside the predetermined limits, paused the PCA infusion because of low respiratory rate, and prompted the nurse or physician to take immediate action. After a respiratory event, the incident was reviewed and the nurses were interviewed by a member of the quality department using a standard set of questions. Patient charts were also reviewed for details about the events, predisposing conditions of the patients, and any incidental findings for each case.



In the first 5 months after implementing the capnography modules, a total of 634 patients received postoperative PCA therapy. Of these, 239 (38%) received hydromorphone and 297 (47%) received morphine. The remaining 98 (15%) were given fentanyl. The average age of the patients was 67.5 years, with the exception of 2 patients younger than age 50 years who were otherwise healthy.

Of the 634 patients receiving PCA therapy, 9 (1.4%) had respiratory depression that required intervention to avert further complications. Of the 9 patients with respiratory events, 6 (67%) patients were receiving hydromorphone; the other 3 (33%) patients were receiving morphine. All 9 patients were also receiving supplemental oxygen.

In 7 (78%) of the 9 patients who had respiratory events, the nursing assessment during the patient's arrival from the postanesthesia care unit to the medical-surgical unit had caused concern about the patient's condition. In 2 (22%) of these cases, the patients were otherwise stable on arrival to the unit.

In 7 (78%) of the events, the patients were not receiving a PCA basal rate and the patients' SpO2 was more than 92%. One patient had an SpO2 of 100% and 1 patient had an SpO2 of 76% at the time that the capnographic monitors alarmed, suggesting a critical respiratory depression.

Table 2
Table 2

All respiratory depression events occurred within the first 24 hrs of initiating PCA therapy, with a mean time of 3.4 hrs, excluding an outlier with 23.5 hrs (Table 2). The median time from initiation of PCA therapy to the onset of respiratory events was 3.5 hrs and the average time was 5.6 hrs. In addition, 1 patient had an event 23.5 hrs after starting PCA. When excluding this outlier patient, the average time to a respiratory depression event was 3.4 hrs.

In all cases, the capnography alarm was the impetus for the nurse to check on and assess these patients; the pulse oximetry monitor had not alarmed. On arrival in the patients' rooms, the nurses found their patients to be unresponsive and took immediate action. In 4 (44%) of the 9 cases, the rapid response team was called to intervene, because of a dramatic change in the patient's condition. In another 3 (33%), the attending physician or a resident was called to the bedside to intervene. In the remaining 2 (22%), the nurse stimulated the patient to breathe and continued to closely monitor the patient. In 4 (44%) of these patients, naloxone was administered to reverse the patient's condition (Table 3). In the remaining cases, PCA was discontinued. These interventions resulted in positive outcomes for all the patients.

Table 3
Table 3

Patient Cases

The following case examples illustrate the advantages of using additional safety monitoring to allow for early detection of subclinical respiratory depression for all patients receiving PCA.

Case 1. PCA therapy was initiated in a 78-year-old man following surgical revision of a left total knee replacement. The patient also had a diagnosis of chronic obstructive pulmonary disease (COPD). He received hydromorphone with no basal rate and a PCA bolus dose of 0.2 mg with a 10-minute lock out. After using the PCA device for approximately 3 hrs, the patient's nurse heard the capnography monitor alarming. On entering the room, the nurse found the patient unresponsive. The PCA had automatically paused, which meant that the patient's respiratory had exceeded the predetermined lower limit of 6 bpm for more than 1 or 2 minutes. A rapid response team was called and 0.4 mg of naloxone was administered to the patient. The patient was awakened, saw many clinicians standing around his bed, and asked "What happened?"

Case 2. A 49-year-old woman was placed on PCA after an elective total abdominal hysterectomy and bilateral salpingo oopherectomy for menometrorrhegia. She had no other significant medical history. She received an initial bolus of 1 mg hydromorphone, followed by a basal rate of 0.2 mg/hr and a PCA bolus dose of 0.2 mg with a 10-minute lock out. Approximately 5 hrs after PCA therapy was initiated, a capnography monitor alarm brought the nurse to the bedside. The safety system had automatically paused the PCA infusion. The nurse assessment found the patient was unresponsive with a respiratory rate of 6 bpm. A rapid response team was called and 0.2 mg of naloxone was administered. The patient responded, and PCA infusion was discontinued. The patient and her family requested that she remain on capnography monitoring overnight.

Case 3. A 64-year-old woman was placed on hydromorphone PCA after a sigmoid colectomy. Her medical history was significant for hypertension and diverticulitis, but without complications. PCA therapy included no basal rate. The PCA bolus dose was 0.3 mg, with a 10-minute lock out. Approximately 2.5 hrs after PCA was started, the capnography monitoring system alarmed because of a low respiratory rate and automatically paused the PCA infusion. The nurse assessed the patient, the attending physician was contacted, and the PCA was discontinued. Shortly after, the patient's respiratory rate returned to 14 bpm.

Case 4. A 78-year-old woman was placed on hydromorphone PCA therapy after inguinal repair. Her medical history was significant for hypertension and hypercholesterolemia. There was no basal rate of PCA, and the bolus rate was 0.5 mg, with a 10-minute lock out. Approximately 2 hrs after PCA therapy was initiated, the patient's respiratory rate fell to 7 bpm, and the SpO2 was 90% while receiving 5 L supplemental oxygen by nasal cannula. The nurse found the patient to be unresponsive. The PCA was discontinued and naloxone was administered. The patient's respiratory rate and level of consciousness returned to baseline within 30 minutes, and her SpO2 returned to 97% within an hour.


The Joint Commission describes pain as the "fifth vital sign" that should be monitored with the same vigilance as blood pressure, pulse, temperature, and respiratory rate. Inadequate pain control is unethical, clinically unsound, and economically wasteful.22 Pain management experts have stated a pain level that is unacceptable to the patient is rated by the patient as a 4 or more on a 0-to-10-point pain scale and requires prompt intervention—obtaining an order for and administration of prescribed analgesic and nonpharmacologic modalities, as appropriate.

The implementation of capnography modules to monitor ventilatory response of patients receiving PCA therapy was an important part of hospital-wide efforts to enhance patient safety and quality of care in postoperative pain management. The American Society of Anesthesiologists emphasizes that ventilation and oxygenation are separate physiologic processes and that monitoring oxygenation by pulse oximetry is not a substitute for monitoring ventilatory function by capnography.23

A growing body of evidence supports capnography's sensitivity in the detection of respiratory depression in patients receiving PCA therapy in general nursing care.12,13 Blinded capnography frequently identified respiratory depression undetected by the treating physicians.14 Another study found that a majority of patients with acute respiratory events had EtCO2 abnormalities that occurred before oxygen desaturation or observed hypoventilation.15 During procedural sedation in children, capnography allowed early detection of arterial oxygen desaturation because of alveolar hypoventilation in the presence of supplemental oxygen.16

The results of this case series provides additional evidence that pulse oximetry may fail to detect respiratory depression, particularly if a patient is receiving supplemental oxygen.24 In all 9 patients who experienced respiratory events, capnography monitors provided the only alert to declining respiratory function. Pulse oximetry did not alarm in any case. In 7 of the 9 cases, the SpO2 was greater than 92%. In the detection of respiratory depression, monitoring respiratory rate and EtCO2 concentration was more effective than monitoring oxygenation alone.

Results underscore the need for vigilance in postoperative care of patients receiving PCA, especially in the first 24 hrs. Another study found that 77.4% of patients suffered respiratory events in the first 24 hrs postoperatively. 25,26 In our series, with excluding one outlier of 23.5 hours, all 9 patients experienced respiratory depression within the first 24 hrs with a mean time of 3.4 hrs.

Factors such as COPD, smoking, and obstructive sleep apnea can significantly increase the risk of respiratory events.27-29 In this case series, 89% of patients who had respiratory events had no history of respiratory disease. This supports the use of capnography to monitor all patients receiving PCA therapy, not only those recognized as being at high risk for opioidinduced respiratory depression.

Several studies have shown that a basal rate increases the total analgesic dose used and risk of respiratory depression.30-34 It is a common belief among clinicians that removing the basal infusion rate will potentially avoid the risks of respiratory depression. However, in our study, only 33% of the patients who required medical intervention were receiving a basal infusion rate. The occurrence of respiratory events in patients receiving no basal rate and having no evidence of PCA by proxy suggests that PCA by demand-bolus alone can deliver sufficient opioid to result in respiratory depression.

Hydromorphone is approximately 8 times more potent than morphine.35 The increased potency allows for better pain control but may increase the risk of respiratory events. In this case series, a much smaller percentage of the 634 patients receiving PCA therapy received hydromorphone than morphine (38% vs 47%), yet a greater percentage of respiratory events was related to hydromorphone than to morphine (67% vs 33%). Other studies have also shown an increased risk of respiratory events with hydromorphone compared with morphine.25

Several studies have shown that a basal rate increases the total analgesic dose used and incidence of side effects such as respiratory depression.30-32 Many clinicians believe that removing the basal infusion rate will potentially avoid the risks of respiratory depression. However, in this case series, 78% of patients experiencing PCArelated respiratory depression had no basal rate.


Since October 2006, we had 9 patients whose level of sedation while receiving PCA opioids led to low respiratory rates. In each case, capnographic monitoring of respiratory rate alerted caregivers before pulse oximetry and prevented a serious adverse event. In most of the patients in this series, nursing assessment had already raised concerns about the patient's condition. Capnographic monitoring provided real-time data that alerted nurses when immediate action was needed. All patients were properly treated without injury.

Concerns about potential respiratory depression also can prevent clinicians from treating pain adequately,10 which can adversely affect recovery and outcomes. The use of continuous capnography to monitor EtCO2 and effective respiratory rate improves clinicians' confidence that opioid doses can be safely increased to provide more effective pain management for all patients.

Other benefits of using capnography to monitor patients receiving PCA therapy include:

  • Increased nursing awareness of changes in respiratory status during PCA therapy
  • Increased early nursing interventions to correct PCA-related respiratory distress
  • Increased nursing confidence that patients were being effectively monitored for respiratory depression
  • Effective use of the rapid response team to correct PCA-related respiratory distress and avoid increased costs of care that would otherwise follow respiratory arrest
  • Measurable increases in patient safety during PCA therapy.


The cases in this series provide additional evidence that continuous capnographic monitoring offers more reliable and earlier indication of changes in patient respiratory status compared with continuous pulse oximetry, particularly in patients receiving supplement oxygen administration. Hospital-defined alerts and data from the capnographic monitoring modules provided clinicians with the earliest warnings of respiratory distress. Automatic pausing of the PCA infusion by the capnography system and expeditious interventions by members of a multidisciplinary healthcare team helped prevent serious complications and the need for more intensive care. Without this PCA safety technology, patients might have needed additional interventions that could have negatively impacted their lives, increased their length of stay, increased the cost of their care, and utilized additional healthcare resources.


  1. Maddox RR, Williams CK, Fields M. Respiratory monitoring in patient-controlled analgesia. Am J Health Syst Pharm. 2004;61:2628,2635.
  2. McPhee SJ, Papadakis MA, Tierney LM Jr, eds. Gonzales R, Zeiger R, online eds. Current Medical Diagnosis and Treatment 2007. San Francisco, CA: McGraw-Hill Medical; 2007.
  3. ISMP Medication Safety Alert. Safety issues with patient-controlled analgesia Part I—How errors occur. July 10, 2003. Accessed May 13, 2008.
  4. ISMP Medication Safety Alert. Frequent problems with medication systems noted during ISMP hospital evaluations Part 2. June 17, 1998. Accessed May 13, 2008.
  5. ISMP Medication Safety Alert. More on avoiding opiate toxicity with PCA by proxy. May 29, 2002. Accessed May 13, 2008.
  6. Etches RC. Respiratory depression associated with patient-controlled analgesia: a review of eight cases. Can J Anaesth. 1994;41:125-132.
  7. JCAHO Sentinel Event Alert. Patient-controlled analgesia by proxy. December 2, 2004. Accessed May 13, 2008.
  8. Overdyk FJ. PCA presents serious risks. APSF Newsl. 2005;20:33.
  9. Cohen MR. How to prevent errors—safety issues with patient controlled analgesia (Part 2). ISMP medication error report analysis. Hosp Pharm. 2005;3:210-212,274.
  10. Bailey P. Supplemental oxygen is appropriate. APSF Newsl. Spring 2006;21:18.
  11. Weinger M. Dangers of Postoperative Opioids; APSF Workshop and White Paper Address Prevention of Postoperative Respiratory Complications. APSF Newsl. Winter 2006;21:61-68.
  12. Maddox R, Williams CK, Oglesby H, et al. Clinical experience with patient-controlled analgesia using continuous respiratory monitoring and a smart infusion system. Am J Health Syst Pharm. 2006;63:157-164.
  13. Hutchison R, Rodriquez L. Capnography and respiratory depression. Am J Nurs. 2008;108:35-39.
  14. Deitch K, Chudnofsky CR, Dominici P. The utility of supplemental oxygen during emergency department procedural sedation and analgesia with midazolam and fentanyl: a randomized, controlled trial. Ann Emerg Med. 2007;49:1-8.
  15. Burton JH, Harrah JD, Germann CA, et al. Does end-tidal carbon dioxide monitoring detect respiratory events prior to current sedation monitoring practices? Acad Emerg Med. 2006;13:500-504.
  16. Lightdale JR, Goldmann DA, Feldman HA, et al. Microstream capnography improves patient monitoring during moderate sedation: a randomized, controlled trial. Pediatrics. 2006;17:e1170-e1178.
  17. Miner JR, Heegaard W, Plummer D. End-tidal carbon dioxide monitoring during procedural sedation. Acad Emerg Med. 2002;9:275-280.
  18. Cashman JN, Dolin SJ. Respiratory and haemodynamic effects of acute postoperative pain management: evidence from published data. Br J Anaesth. 2004;93:212-223.
  19. Walder B, Schafer M, Henzi I, et al. Efficacy and safety of patientcontrolled opioid analgesia for acute postoperative pain. A quantitative systematic review. Acta Anaesthesiol Scand. 2001;45:795-804.
  20. Overdyk FJ, Carter R, Maddox RR, et al. Continuous oximetry/capnometry monitoring reveals frequent desaturation and bradypnea during patient-controlled analgesia. Anesth Analg. 2007;105:412-418.
  21. Colman Y, Krauss B. Microstream capnography technology: a new approach to an old problem. J Clin Monitor Comput. 1999;15:403-409.
  22. Phillips DM. JCAHO pain management standards are unveiled. Joint Commission of Accreditation of Healthcare Organizations. JAMA. 2000;284:428-429.
  23. American Society of Anesthesiologists Task Force on Sedation and Analgesia by Non-Anesthesiologists. Practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology. 2002;96:1004-1017.
  24. Poirier MP, Gonzalez Del-Rey JA, McAneney CM, et al. Utility of monitoring capnography, pulse oximetry, and vital signs in the detection of airway mishaps: a hyperoxemic animal model. Am J Emerg Med. 1998;16:350-352.
  25. Taylor S, Kirton O, Staff I, et al. Postoperative day one: a high risk period for respiratory events. Am J Surg. 2005;190:752-756.
  26. Shapiro A, Zohar E, Zaslansky R, et al. The frequency and timing of respiratory depression in 1524 postoperative patients treated with systemic or neuraxial morphine. J Clin Anesth. 2005;17:537-542.
  27. Arozullah AM, Conde MV, Lawrence VA. Preoperative evaluation for postoperative pulmonary complications. Med Clin North Am. 2003;87:153-173.
  28. Parikh SN, Stuchin SA, Maca C, et al. Sleep apnea syndrome in patients undergoing total joint arthroplasty. J Arthoplasty. 2002;17:635-642.
  29. Gupta RM, Parvizi J, Hanssen A, et al. Postoperative complications in patients with obstructive sleep apnea syndrome undergoing hip or knee replacement: a case-control study. Mayo Clin Proc. 2001;76:897-905.
  30. Hansen LA, Noyes MA, Lehman ME. Evaluation of patient-controlled analgesia (PCA) verses PCA plus continuous infusion in postoperative cancer patients. J Pain Symptom Manage. 1991;6:4-14.
  31. Smythe MA, Zak MB, O'Donnell MP, et al. Patient-controlled analgesia versus patient-controlled analgesia plus continuous infusion after hip replacement surgery. Ann Pharmacother. 1996;30:224-227.
  32. Looi-Lyons LC, Chung FF, Chan VW, et al. Respiratory depression: an adverse outcome during patient-controlled analgesia therapy. J Clin Anesth. 1996;8:151-156.
  33. Parker RK, Holtmann B, White PF. Effects of a nighttime opioid infusion with PCA therapy on patient comfort and analgesic requirements after abdominal hysterectomy. Anesthesiology. 1992;76:362-367.
  34. Dal D, Kanbak M, Caglar M. A background infusion of morphine does not enhance postoperative analgesia after cardiac surgery. Can J Anaesth. 2003;50:476-479.
  35. Li JM. Pain management in the hospitalized patient. Med Clin North Am. 2002;86:771-795.
Related Items
Comparing Pharmacy Benefit Managers: Moving Well Beyond the Simple Spreadsheet Analysis
David Calabrese, RPh, MHP
June 2008, Vol 1, No 5 published on July 21, 2010 in Business
Last modified: November 10, 2011
  •  Association for Value-Based Cancer Care
  • Value-Based Cancer Care
  • Value-Based Care in Rheumatology
  • Oncology Practice Management
  • Rheumatology Practice Management
  • Urology Practice Management
  • Inside Pharmacy
  • Lynx CME