In December 1999, a 19-year-old woman entered the hospital for a routine same-day surgery. The woman's postoperative pain was difficult to manage, so the surgeon admitted her. The assigned surgical RN was surprised to read a medication order for "I.V. push morphine 4 to 8 mg q 15 minutes - 2 hours PRN." The dose seemed too high, so the RN decided she wouldn't give more than 4 to 6 mg at a time.
The RN recorded the patient's blood pressure, pulse, and respirations as prescribed by hospital policy. The documentation sheet included areas for vital signs, but not for a pain or sedation score. Pain and sedation scores were required documentation for patients on patient-controlled analgesia therapy or continuous narcotic infusions, but not I.V. push narcotics. The facility's med-surg units had just started administering I.V. push narcotics; previously, only critical care areas had administered these "high-alert" drugs.
The patient was crying, restless, and frequently requesting pain medication. After 6 hours, the RN had given the patient a total of 32 mg of I.V. morphine. The nurse checked on her patient after the final dose and found her sleeping soundly, snoring lightly, and positioned on her abdomen. Although it was unusual to find a postoperative patient lying on his or her stomach, the patient's mother reported it as her daughter's preferred sleeping position.
Forty-five minutes later, the nursing assistant found the patient unresponsive with cyanotic lips, still lying on her stomach. While her parents watched in horror, a code-blue team arrived to resuscitate her, but its efforts were unsuccessful. The patient died from an apparent narcotic-associated respiratory depression.
The nurse was devastated. She repeatedly reviewed her nursing care in her mind, trying to uncover what she missed. She had followed the current policy for recording postoperative vital signs, checked on the patient more frequently than hospital policy required, and administered no more than 4 to 6 mg of I.V. morphine at a time, actually less than what was ordered. The autopsy found nothing that would have contributed to the patient's death.
Organizational response
This tragic case triggered a chart review of all patients who had received naloxone over a 2-month period. The review identified 11 patients who had experienced severe respiratory depression that required naloxone to reverse the oversedation. A multidisciplinary group was chartered in April 2000 to investigate the reasons for respiratory oversedation, create an action plan to correct any contributing factors, and ultimately decrease the incidence of severe narcotic oversedation by 75%. The group consisted of nurses from the surgical unit and postanesthesia care unit (PACU), pharmacists, a house physician, anesthesiologists, respiratory therapists, and a member of the clinical quality department, who served as the group's facilitator. The group was led by a clinical nurse specialist and sponsored by the vice president for medical affairs.
The group's first task was to explore possible root causes of oversedation. Members completed a flowchart of the surgical process, starting with patient arrival to the operating room and ending with patient discharge. Several members felt there was only one primary cause of oversedation-and it occurred after the patient arrived on the surgical unit. Some members were concerned with only their department's actions and missed seeing the whole picture of the patient's surgical experience.
The flowchart helped change their parochial thinking by highlighting times during the surgical process when the processes weren't clear and errors could occur. The flowchart also demonstrated the interdependence between the OR, PACU, and surgical unit, and how an action upstream in the surgical process could impact patient outcomes later in the process.
The group measured progress by reviewing all instances when a patient required naloxone to reverse respiratory sedation, and classified the severity of oversedation. No severity scale was found during a literature review, so the group developed one (Table 1). The measure of progress was the rate of class 3 and 4 events tracked monthly. Later, these rates were annualized and reframed as the number of discharges for every severe episode of oversedation.
Over the next 12 months the group tested and implemented 34 changes (Table 2). By April 2001 the group concluded that it had achieved its goal, so it disbanded (Figure 1). However, later in 2001 the rate of severe narcotic-induced oversedation increased, returning to the same rate as before interventions. The group was reconvened with a goal of once again reducing severe oversedation by 75%. A pain management team, consisting of a 0.8 full-time equivalent (FTE) clinical nurse specialist and a 0.5 FTE clinical pharmacist specialist, was also piloted at this time. This team reviewed the care of all postoperative patients, offered consultation to nurses and physicians, performed extensive staff education, reviewed every case when a patient received naloxone, and implemented 23 additional changes (Table 3).
Results since late 2002 have been dramatic. From 2000 through 2002, Fairview Southdale Hospital, a 350-bed community hospital in Minneapolis, Minn., averaged nine class 3 and 4 events per year. This decreased to 1.7 events per year from 2003 through 2006: a reduction of 81.1% (Figure 2).
These results convinced Fairview Health Services to set a goal for its seven-hospital system to reduce its rate of serious narcotic oversedation by 75%. The corporate director of clinical pharmacy services was assigned to lead this spread effort. In 2004 a system-wide Pain Committee was established as a subcommittee of the Fairview Formulary and Drug Use Committee. Chaired by a clinical pharmacy pain specialist, this committee is responsible for refinement of tools, the spread of best practices, development of relevant computer decision support, and other aspects of safe narcotic use. (See Table 4 for a list of actions taken at a system level.)
Campaign participation
Since 2004, the annual number of serious narcotic oversedation cases at Fairview has fallen from 25 to 10, a 60% reduction. By continuing to focus on reducing narcotic oversedation, Fairview is actively contributing to the 5 Million Lives Campaign, created by the Institute for Healthcare Improvement (IHI). According to the IHI, 58% of medication injuries are due to high-alert medications. The four high-alert medications most responsible for injuries are anticoagulants, sedatives, narcotics, and insulin. The general principles for reducing harm from high-alert medications are prevention, identification, and mitigation.1
The majority of Fairview's actions in reducing severe oversedation follow the above principles. Interventions range from using standard order sets for certain pain medications to taking practical precautions to prevent potential errors in dispensing and administration. For example, the facility's pharmacists repackage hydromorphone, I.V. pain medication, from 2-mL syringes into 0.2 mg, the usual dosage.2
From tragedy to teaching
Fairview Southdale Hospital supported the surgical RN who provided care to the deceased patient. The nurse wasn't fired and didn't lose her license. She took an active role in redesigning the policies for the care of postoperative patients. The postoperative documentation now includes areas to chart a patient's pain and sedation level, and respiratory rate, depth, and quality-regardless of the route of the narcotic. Postoperative pain orders list recommended dose ranges for I.V. narcotics, and physicians are contacted when narcotics are ordered in unusually high doses.
Nurses at the facility receive this case study as a learning tool. The nurse involved often retells this patient's story, urging her peers to question high narcotic doses and thoroughly document patients' responses. She shares her experience in the pursuit of safe patient care and reduction of harm from high-alert medications.
REFERENCES