Membership information 0800 225 677
Medicolegal advice 0800 982 766

Respiratory depression from postoperative analgesia

01 September 2006

Mr K, a 37-year-old professional, was admitted for an elective anterior resection of the bowel. Dr A, a consultant anaesthetist, performed the anaesthetic, inserting an epidural catheter to manage postoperative pain with a continuous infusion.

The infusion mixture contained bupivacaine and fentanyl. He also prescribed “rescue analgesia” in case the epidural failed.

Postoperative care was delegated to the nursing staff and junior doctor, Dr B. The epidural worked well for 24 hours but failed shortly before midnight. In accordance with Dr A’s instructions, Dr B administered 20mg of morphine intramuscularly and commenced patient-controlled analgesia (PCA). This was not effective, so Dr B administered a further 10mg of morphine, this time intravenously. He had also prescribed a dose of zopilcone earlier. The nursing notes indicate that a “background infusion” was commenced on the PCA pump.

Dr B left the patient shortly after this, but gave no specific instructions about monitoring. No further observations were recorded until about an hour later. The nurses noted that the patient’s oxygen saturation (value not stated) had dropped and he was unresponsive; they called Dr B. At least 48mg of morphine had been administered to the patient in the preceding two hours.

Dr B found that the airway was obstructed. The oxygen saturation at this time was about 60%. Dr B relieved the obstruction, gave oxygen and administered naloxone. He telephoned Dr A, who arrived shortly thereafter to manage the patient. The patient survived the incident, but suffered hypoxic brain damage in the form of loss of memory and cognitive functions.

Expert opinion

A consultant anaesthetist concluded that the large dose of morphine had indeed caused the respiratory depression and severe hypoxia. This could have been prevented if he had been adequately monitored.

The expert felt that the overall standard of care provided by Dr A was, at all stages, acceptable.

It was reasonable to assume that the nurses would carry out the appropriate postoperative observations without any further instructions.

He was highly critical of the nursing staff; the “frequency of observations was hopelessly inadequate” and there was no evidence of a protocol for observations of patients on PCA.

A second consultant anaesthetist and a specialist respiratory physician were critical of Dr B’s actions. They felt he should have recognised that the intravenous dose of 10mg morphine shortly after the 20mg intramuscular dose was hazardous. The effects might have been compounded by the fentanyl and zoplicone.

Dr B should have given smaller increments of the intravenous dose and stayed with the patient for a longer period to observe him. He should also have discussed suitable observations with the nurses and written up the instructions.

The case was considered indefensible and was settled out of court for £375,000 (US$708,000), of which the hospital contributed 20%.

Learning points

  • Modern, complex techniques for postoperative pain management require appropriately trained staff and adequate facilities.  
  • All healthcare facilities using PCA should have an established protocol, including how to set up a PCAS for the patient; suggested dose regimens and the necessary observation standards.   
  • The incidence of opioid-induced respiratory depression with PCA ranges between 0.1–0.8%.The addition of a background infusion increases this complication to between 1.1–3.9% but contributes little, if any, improvement in the analgesia. The routine use of a background infusion is therefore generally not recommended but if it is prescribed, the best setting for the patient is the High Dependency Unit, the Intensive Care Unit or the recovery suite.   
  • The most common route of delivery is intravenously. No additional opiate should be given, either intravenously (IV) or intramuscularly (IM).Drug absorption from the IM route is variable, and the patient may suddenly experience a bolus effect.

Further information

  • Pain Management Services – Good Practice, Royal College of Anaesthetists and The Pain Society (May 2003).
  • Macintyre P E, Safety and Efficacy of Patient-Controlled Analgesia,Br J Anaesth 87 (1):36–46 (2001).
  • Stone M and Wheatley B, Patient-ControlledAnalgesia, Br J Anaesth CEPD Reviews 2 (3):79–82 .