Procedural Sedation in the Emergency Department - An Observational Study: Does Nil Per Os Status Matter?

Abstract

Introduction: Procedural sedation (PS) is commonly performed in the emergency department (ED). Nil per os (nothing by mouth) (NPO) guidelines extrapolated from standards for patients undergoing elective procedures in the operating room have been applied to ED PS patients. There has been no large study of ED PS patients comparing differences in adverse events and PS success rates based on NPO status.

Methods: From a cohort of consecutive ED PS patients of all ages in the 20 EDs of one hospital system-one quaternary ED, four tertiary EDs, six community hospital EDs, one rural ED, two pediatric EDs, and six freestanding EDs in two states in the Midwest and South-we conducted a retrospective analysis on a prospective database over 183 months from April 2000-June 2015. Primary outcome was the incidence of side effects and complications, which comprised the adverse effects. The side effects were nausea, vomiting, itching/rash, emergence reaction, myoclonus, paradoxical reaction, cough, and hiccups. Complications were oxygen desaturation <90%, respiratory depression (respiratory rate <8), apnea, tachypnea, hypotension, hypertension, bradycardia, and tachycardia. Normal vital signs were age dependent. Secondary outcome was successful sedation defined as completion of the procedure. We examined the association between adverse events and successful sedation with NPO status.

Results: Of 3,274 visits, exact NPO status was known in 2,643 visits. Comparison of NPO <8 hours in 1,388 patients vs ≥ 8 hours in 1,255 patients revealed side effects 5.5% vs 4.5% (P = 0.28); complications 11.9% vs 17.7% (P < 0.001); adverse events 16.3% vs 21.5% (P < 0.001), interventions 4.1% vs 4.4% (P = 0.73), and procedural completions 94.3% vs 89.7% (P < 0.001). After adjustment for age, sex, transfer status, American Society of Anesthesiology physical status classification, race, primary sedative, multiple sedatives, sedative plus analgesic, and primary analgesic, we found no association between NPO status and side effects (P = 0.68), complications (P = 0.48), or adverse effects (P = 0.26); however, procedural completion rate remained significantly higher for NPO < 8 hours (P = 0.007).

Conclusion: A nil per os status ≥8 hours may have similar or worse outcomes than NPO <8 hours, which is contrary to many suggested guidelines. Strict adherence to NPO guidelines in ED procedural sedation patients may not be necessary.