206 Safety of Single- vs Two-physician Procedural Sedation in a Small Community Emergency Department

Abstract

Sedation is commonly required for painful procedures in the emergency department (ED). Many hospital have policies that mandate two physicians participate in deep sedation cases. Evidence is lacking, however, that a two-physician approach improves safety outcomes. We compared the safety of ED procedural sedation between a two-physician and a single-physician policy in a small, single-coverage community ED. This is a before/after single-center observational study of prospectively collected data from January 2013 through December 2016. The study population included a consecutive series of ED patients of any age who received procedural sedation for any indication during the study period. Data collection occurred within the context of a facility-wide quality assurance and process improvement program that included standardization of sedation data reporting and monthly structured safety audits of all procedural sedation cases in the medical center. In 2012 a policy was implemented which required two physicians to participate in deep sedations in the ED. Given the practice environment in this small single-coverage community ED, it was frequently impractical to have two physicians reliably available for time-sensitive urgent, and often emergent, procedures in the ED. This led to impaired delivery of modern, high-quality ED sedation services, most notably delays in patient care. In September 2014, our medical center switched from a two-physician policy to a single-physician policy requiring only one emergency physician, accompanied by a sedation-trained ED registered nurse, and often a respiratory therapist. The primary outcome was a sedation-related escalation of care that resulted in one of the following adverse events: dysrhythmia (symptomatic bradycardia or ventricular arrhythmias), cardiac arrest, endotracheal intubation, or unanticipated hospitalization. Secondary outcomes included hypoxemia (peripheral oxygen saturation less than 90% for greater than one minute), the need for bag-valve mask ventilation (BVM), use of a reversal agent, laryngospasm or pulmonary aspiration. We performed 381 sedations during the study period: 135 patients in the two-physician group (before) and 246 patients in the single-physician group (after). The two groups were comparable in age and sex. Procedures for which sedation was indicated were similar between the groups, with joint or fracture reductions being the most common. Deep and dissociative sedation was more commonly employed in both groups, but was significantly more common in the single-physician group than the two-physician group (93% vs 68%, p<0.001). Propofol and ketamine were the most commonly used sedative agents for deep sedation in both groups. There was no occurrence of the primary outcome, and no clinically relevant adverse events in either group. Secondary outcomes were uncommon, and were similar in the two groups. The use of BVM was 1.5% in the two-physician group and 3.2% in the single-physician group (p=0.5). Rates of hypoxemia and other expected sedation events were uncommon (0-1.5%) and were similar in both groups. In this small, single-coverage community ED, replacement of a two-physician policy with a single-physician policy for deep sedation in the ED was not associated with an increase in adverse events.