Risk factors leading to failed procedural sedation in children outside the operating room

Deep sedation enables effective performance of imaging or procedures in children, but failed sedation still occurs. We desired to determine the factors that were associated with failed sedation in children receiving deep sedation by a dedicated nonanesthesia sedation service and hypothesized that the presence of an upper respiratory infection (URI) and/or other risk factors would increase the probability of failing sedation.

Patient sedation records from January 2007 to December 2011 were reviewed to identify 83 failed sedations. A convenience sample of 523 patients with successful sedation from January 2009 to February 2009 was identified for comparison.

Seven of the 13 predictors were significantly associated with failed sedation; these are as follows: (1) URI (P = 0.008); (2) congenital heart disease (P = 0.021); (3) obstructive sleep apnea (OSA)/snoring (P < 0.001); (4) the American Society of Anesthesiologists (ASA) class of above II (P < 0.001); (5) obesity (P < 0.001); (6) increased weight (P < 0.001); and (7) older age (P < 0.001). Sex, prematurity, asthma, gastroesophageal reflux, and cerebral palsy/developmental delay were not associated with failure. Pulmonary hypertension was not able to be assessed because only 1 patient with pulmonary hypertension was sedated. A forward stepwise regression identified 5 variables that could be considered useful predictors of failed sedation, which are as follows: (1) URI (odds ratio [OR], 2.73 [range, 1.58-4.73]); (2) OSA/snoring (OR, 2.06 [range, 1.22-3.48]); (3) ASA class III (OR, 2.31 [range, 1.40-3.84]); (4) obesity (OR, 1.95 [range, 1.01-3.75]); and (5) older age (OR, 1.15 [range, 1.08-1.21).

Presence of a URI, a history of OSA/snoring, ASA class III, obesity, and older age are associated with increased probability of failed sedation. A prospective, multicenter observational study would allow for the robust modeling of comorbidities to guide pediatric sedation management.