Critical respiratory events in the postanesthesia care unit. Patient, surgical, and anesthetic factors.

BACKGROUND: Previous studies have noted a high incidence of adverse outcomes in the postanesthesia care unit (PACU), but few have examined associated factors and patient outcomes. To determine the frequency of acute, unanticipated respiratory problems and to examine the associated patient, surgical, and anesthetic factors, we prospectively collected preoperative, intraoperative, and postoperative data on 24,157 consecutive PACU patients who received a general anesthetic during a 33-month period.
METHODS: A PACU critical respiratory event (CRE), was defined as any unanticipated hypoxemia (hemoglobin oxygen saturation < 90%), hypoventilation (respiratory rate < 8 breaths/min or arterial carbon dioxide tension > 50 mmHg) or upper-airway obstruction (stridor or laryngospasm) requiring an active and specific intervention (ventilation, tracheal intubation, opioid or muscle relaxant antagonism, insertion of oral/nasal airway or airway manipulation). These problems were documented by PACU nurses whereas data on case-mix, surgical factors, and intraoperative management were retrieved from the anesthetic record. Significant patient, surgical, and anesthetic factors were identified by logistic regression analysis. Other morbidity experienced by patients with a CRE was also noted.
RESULTS: For patients given general anesthesia the risk of a CRE was 1.3% (hypoxemia 0.9%, hypoventilation 0.2%, airway obstruction 0.2%). Preoperative factors that increase risk were age > 60 yr, male gender, diabetes, and obesity (P < 0.05). Patients who underwent operative procedures on an emergency basis and whose operation was longer than 4 h were also at increased risk, but those undergoing perineal procedures were at lower risk (P < 0.05). Anesthetic risk factors (P < 0.05) included opioid premedication (relative odds 1.8), sedatives preoperatively (2.0), fentanyl > 2.0 micrograms.kg-1.h-1 as the sole opioid (1.9), fentanyl used in combination with morphine (1.6) and atracurium > or = 0.25 mg.kg-1.h-1 (2.2). Patients in whom anesthesia was induced with thiopental (relative odds 2.5), compared with those who received propofol for induction, were also at increased risk of a CRE. Patients with a CRE stayed longer in PACU, had higher rates of unanticipated admissions to the intensive care unit and were more likely to have PACU cardiac problems (P < 0.01).
CONCLUSIONS: A CRE is relatively rare. Multiple patient and surgical factors and specific aspects of anesthetic management are associated with the occurrence of a CRE in the PACU.