Characteristics of and Predictors for Apnea and Clinical Interventions During Procedural Sedation.

Abstract

We describe the characteristics of and predictors for apnea and clinical interventions during emergency department (ED) procedural sedation. High-resolution data were collected prospectively, using a convenience sample of ED patients undergoing propofol or ketofol sedation. End tidal CO₂ (etco₂), respiratory rate, pulse rate, and SpO₂ were electronically recorded in 1-second intervals. Procedure times, drug delivery, and interventions were electronically annotated. Kaplan-Meier curves were used to describe the onset of clinical interventions as a function of sedation time. The onset of apnea (15 consecutive seconds with carbon dioxide ≤10 mm Hg) and clinical interventions were estimated with a series of Cox proportional hazards survival models, with time to first apnea or clinical intervention as the dependent variable. Finally, we tested the association between apnea and clinical intervention. Three hundred twelve patients were analyzed (53% male patients). Apnea was preceded by etco₂ less than 30 mm Hg or greater than 50 mm Hg at 30, 60, and 90 seconds before its onset. Clinical interventions were predicted by apnea, SpO₂, and propofol use. Increasing age predicted both apnea and interventions. Apnea was not predicted by respiratory rate or SpO₂. Apnea occurred in half of the patients and clinical interventions in a quarter of them. Clinical intervention was not predicted by abnormal respiratory rate or abnormal etco₂ level. The majority of clinical interventions (85%) were minor, with no cases of assisted ventilation, intubation, or complications. Alterations in etco₂ predicted apnea along a specific time course. Alterations in SpO₂, apnea, and propofol use predicted clinical interventions. Increasing age predicted both apnea and clinical intervention.