Patient safety in the recovery room

Adverse sedation events in pediatrics: a critical incident analysis of contributing factors

OBJECTIVE

Factors that contribute to adverse sedation events in children undergoing procedures were examined using the technique of critical incident analysis.

METHODOLOGY

We developed a database that consists of descriptions of adverse sedation events derived from the Food and Drug Administration's adverse drug event reporting system, from the US Pharmacopeia, and from a survey of pediatric specialists. One hundred eighteen reports were reviewed for factors that may have contributed to the adverse sedation event. The outcome, ranging in severity from death to no harm, was noted. Individual reports were first examined separately by 4 physicians trained in pediatric anesthesiology, pediatric critical care medicine, or pediatric emergency medicine. Only reports for which all 4 reviewers agreed on the contributing factors and outcome were included in the final analysis.

RESULTS

Of the 95 incidents with consensus agreement on the contributing factors, 51 resulted in death, 9 in permanent neurologic injury, 21 in prolonged hospitalization without injury, and in 14 there was no harm. Patients receiving sedation in nonhospital-based settings compared with hospital-based settings were older and healthier. The venue of sedation was not associated with the incidence of presenting respiratory events (eg, desaturation, apnea, laryngospasm, approximately 80% in each venue) but more cardiac arrests occurred as the second (53.6% vs 14%) and third events (25% vs 7%) in nonhospital-based facilities. Inadequate resuscitation was rated as being a determinant of adverse outcome more frequently in nonhospital-based events (57.1% vs 2.3%). Death and permanent neurologic injury occurred more frequently in nonhospital-based facilities (92.8% vs 37.2%). Successful outcome (prolonged hospitalization without injury or no harm) was associated with the use of pulse oximetry compared with a lack of any documented monitoring that was associated with unsuccessful outcome (death or permanent neurologic injury). In addition, pulse oximetry monitoring of patients sedated in hospitals was uniformly associated with successful outcomes whereas in the nonhospital-based venue, 4 out of 5 suffered adverse outcomes. Adverse outcomes despite the benefit of an early warning regarding oxygenation likely reflect lack of skill in assessment and in the use of appropriate interventions, ie, a failure to rescue the patient.

CONCLUSIONS

This study-a critical incident analysis-identifies several features associated with adverse sedation events and poor outcome. There were differences in outcomes for venue: adverse outcomes (permanent neurologic injury or death) occurred more frequently in a nonhospital-based facility, whereas successful outcomes (prolonged hospitalization or no harm) occurred more frequently in a hospital-based setting. Inadequate resuscitation was more often associated with a nonhospital-based setting. Inadequate and inconsistent physiologic monitoring (particularly failure to use or respond appropriately to pulse oximetry) was another major factor contributing to poor outcome in all venues. Other issues rated by the reviewers were: inadequate presedation medical evaluation, lack of an independent observer, medication errors, and inadequate recovery procedures. Uniform, specialty-independent guidelines for monitoring children during and after sedation are essential. Age and size-appropriate equipment and medications for resuscitation should be immediately available regardless of the location where the child is sedated. All health care providers who sedate children, regardless of practice venue, should have advanced airway assessment and management training and be skilled in the resuscitation of infants and children so that they can successfully rescue their patient should an adverse sedation event occur.

The Australian Incident Monitoring Study. Recovery room incidents in the first 2000 incident reports

Of the first 2000 incidents reported to the Australian Incident Monitoring Study 120 (6%) occurred in the recovery room after general, regional or local anaesthesia. Over two thirds (69%) of these involved the respiratory system, 19% were cardiovascular, 3% involved the central nervous system and 9% were miscellaneous in nature. These recovery room incidents were associated with significantly more adverse outcomes (56%) than incidents in the operating theatre (24%). The types and relative frequencies of these recovery room incidents were similar to those of serious recovery complications in a recent analysis of closed malpractice claims; this suggests that incident monitoring may be useful in the study and prevention of recovery room complications. Over three quarters (77%) of all recovery incidents (and 88% of respiratory incidents) were detected clinically; the remainder were first detected by a monitor. A theoretical analysis showed that over 95% of respiratory events, had they been allowed to evolve, would have been detected by pulse oximetry before organ damage occurred, emphasising the potential importance of pulse oximetry in reducing adverse outcome from any complication in the recovery ward which might be "missed" by clinical observation. The findings of this study underline the importance of having an adequate number of trained recovery nursing staff supported by the availability of a pulse oximeter for each patient at least until the return of protective reflexes and the ability to maintain adequate arterial saturation has been established.

Recovery from anaesthesia in children

There are no published comprehensive surveys of paediatric recovery room experience and the incidence of complications. A prospective survey was made of 16,700 consecutive admissions to the recovery room at the Royal Manchester Children's Hospital during the years 1985-1988. The incidence of respiratory complications was low, with laryngospasm 0.85%. The incidence of hypotension was higher than that in adult studies; over 50% of children recorded a decrease in blood pressure in the recovery room of more than 20%, compared to values before operation. The incidence of vomiting in the recovery room was also lower than in comparable adult studies. Certain aspects of recovery room practice changed during the 4 years of the study; these included routine oxygen administration, parents in the recovery room, and our approach to postoperative analgesia. The implications of these changes are discussed.

A prospective survey of anaesthetic critical events in a teaching hospital

A survey of anaesthetic critical events based on voluntary anonymous reports was undertaken over a two and a half year period within a teaching hospital Department of Anaesthesia. At the end of the survey 167 reports were analysed. Human error was a contributing factor in 82% of events. Inadequate preoperative patient assessment or preparation, problems in the area of human/equipment interface and various stress factors for the anaesthetist featured significantly in the survey. The method enables the collection of objective data on factors contributing to anaesthetic and surgical risk and the formulation and evaluation of potential corrective strategies. It also facilitates harmonious peer review via individual and group feedback activities. The adoption of such a survey on a wider scale is seen as a valuable part of quality assurance in the continuing attempt to increase patient safety.