Circulation 75,648 • Volume 20, No. 2 • Summer 2005   Issue PDF

PCA Presents Serious Risks

Mark R. Rosekind, PhD

To the Editor

A healthy 26-year-old soccer player undergoes cruciate ligament repair and is discharged to the floor with patient-controlled analgesia (PCA). His mother, concerned that he rest comfortably after poor postoperative pain control, repeatedly presses his morphine PCA button while he is asleep. He stops breathing and is resuscitated, albeit with a hypoxic brain injury.

A mother of 3 small children with chronic lower back pain, for which she takes Percocet™, undergoes uneventful emergency appendectomy and is discharged to the floor on hydromorphone PCA. Frustrated by her difficult pain management, floor nurses give supplemental intravenous hydromorphone outside of PCA parameters. She suffers a respiratory arrest and hypoxic brain injury.

Almost 20 years after the APSF spearheaded the adoption of pulse oximetry and capnography as standards for patient monitoring in the OR,1 and with a resultant decrease in catastrophic respiratory events,2 I still encounter such cases as an expert witness in defense of hospitals liable for these devastating outcomes.

PCA by Proxy Presents Great Danger

There is accumulating evidence that PCA presents serious risks to patients and liability exposure to hospitals. The 2 main safety mechanisms integrated in current PCA pumps, the lock-out interval and maximum dose limit, fall woefully short of preventing overdosage of narcotics. PCA “by proxy,” in which family members or persons other than the patient administer bolus doses, as occurred in the example above, are well documented.3 The risk of patient harm due to medication errors with PCA pumps, the most common form of PCA error, is 3.5-times the risk of harm to a patient from any other type of medication administration error.4 The FDA’s Manufacturer and User Facility Device Experience (MAUDE) Database for 2004, a voluntary database for reporting problems with devices, reports 21 deaths related to PCA pumps versus 16 deaths for all other large volume infusion pumps (LVP).5 The installed number of LVPs is approximately 10-times greater than the number of PCA pumps, suggesting the risk of death from an adverse event with a PCA pump is at least 10-times greater than with LVPs. Recent University Hospital Consortium (UHC) survey data found that 9 of 15 members who responded to a survey identified recent PCA adverse events as potentially costly in terms of liability exposure.

The 2004 ASA Annual Meeting (www.asaabstracts.com) featured many “better mousetraps” for pulse oximetry, capnography, and systems to prevent and intervene in respiratory insufficiency. Yet somehow, all this technology stops at the post-anesthesia care unit (PACU) door, and patients are sent to the hospital floor with their narcotic laden infusions, where currently acceptable standards of monitoring are a flow sheet with vital signs and pump data that are manually updated at best every 30 minutes. The patient’s only safety net is the mayday team, an often poorly choreographed band of practitioners left to initiate resuscitation at an often undetermined period of time after the critical event.

In the mid 1990s, management of acute postoperative pain and PCA was, in effect, transferred to the surgical service when the government (HCFA/CMS) and private insurers stopped compensating anesthesiologists for managing postoperative pain. It was left to the surgeons to manage pain, narcotics, and respiratory insufficiency. Yet if you asked first-year medical students which specialist is the expert in pain, narcotics, respiratory physiology, and resuscitation, they would reply “anesthesiologist,” not “surgeon,” “internist,” or “hospitalist.” We define ourselves as perioperative physicians,6 yet in the ever increasing number of hospitals without an anesthesia-run acute pain service, patients are discharged from the PACU with PCA technology that exposes them to great risk without our oversight.

Capnography May Warn of Hyperventilation

Some may argue that my experience lacks a denominator, and that these may be rare events. I contend that one such adverse outcome in a friend or family member undergoing elective surgery is one too many. Cashman et al., in a recent meta analysis of the literature, reported an incidence of “respiratory depression” of 11.5% by oximetry (O2Sat <90%) and 1.3% by bradypnea (RR <10), which translates to thousands of patients with potentially catastrophic respiratory depression per day.7 Others take comfort in using oximeters in “high-risk” patients, yet this is a deceptively ineffective approach. The first patient above could hardly be described as “high risk.” Furthermore, Stemp recently highlighted the dangers of simply adding supplemental oxygen, the typical “therapeutic” response to desaturation.8 Supplemental oxygen does not treat desaturation due to hypoventilation, but merely postpones the patient’s insidious progress from bradypnea to apnea. The addition of a capnograph, however, may anticipate a patient’s desaturation by warning of a decrease in respiratory rate and rise in end-tidal carbon dioxide. In the OR setting, these monitors together were judged to potentially prevent 93% of respiratory mishaps in the closed claims database.2

Perhaps the APSF could take a more active role in measuring the scope of the problem, propose monitoring guidelines for PCA, and encourage development of technology that closes the monitoring void that exists with PCA. It would once again demonstrate that it is the premier organization devoted to patient safety in the perioperative environment.

Frank J. Overdyk, MSEE, MD
Charleston, SC


References

  1. Pierce EC. The establishment of the APSF and the ASA Closed Claims Study. The 34th Rovenstine Lecture. Available on the web at: https://www.apsf.org/about/rovenstine/part1.mspx. Accessed on April 26, 2005.
  2. Tinker J, Dull D, Caplan RA, et al. Role of monitoring devices in prevention of anesthetic mishaps: a closed claims analysis. Anesthesiology 1989;71:541-6.
  3. Sidebotham D, Dijkhuizen M, Schug S. The safety and utilization of patient-controlled analgesia. J Pain Symptom Manage 1997;14:202-9.
  4. Sullivan M, Phillips MS, Schneider P. Patient-controlled analgesia pumps. USP Quality Review 2004;81:1-3. Available on the web at: http://www.usp.org/ pdf/patientSafety/qr812004-09-01.pdf. Accessed April 25, 2005.
  5. Manufacturer and User Facility Device Experience Database – (MAUDE). U.S. Food and Drug Administration, Center for Devices and Radiologic Health, Department of Health and Human Services. Available on the web at: http://www.fda.gov/cdrh/maude.html.
  6. Alpert CC, Conroy JM, Roy RC. Anesthesia and perioperative medicine: a department of anesthesiology changes its name. Anesthesiology 1996;84:712-5.
  7. Cashman JN, Dolin SJ. Respiratory and haemodynamic effects of acute postoperative pain management: evidence from published data. Br J Anaesth 2004;93:212-23.
  8. Stemp L. Etiology of hypoxemia often overlooked. APSF Newsletter 2004;19(3): 38-9.