Apnea Risk Stratification in Emergency Medicine (Patient Safety)

November 2008

Apnea Risk Stratification in Emergency Medicine (Patient Safety)

Search Strategy: Using PUBMED Clinical Queries, you select a narrow/specific search for clinical prediction guides using the search term “apnea” yielding 177 citations on September 29, 2008. From this list you select the papers on the STOP questionnaire derivation and the Berlin Questionnaire validation. Your Anesthesiology colleagues provide you with two additional references which are not obtained on any search using PUBMED clinical queries for diagnosis, prognosis, or clinical prediction guide and the search term “apnea”.

Working the back pod with a procedurally disinclined off-service resident, you note the arrival of a 400-pound female who fell in the driveway while walking to retrieve the newspaper on an icy February morning. EMS reports that despite 4 mg then 10 mg Morphine bolus doses, she is screaming in pain. They didn’t need to inform you about her screaming because everyone from EM-1 through EM-33 has heard the cries of pain from the moment she arrived. You quickly ascertain that Morphine, as currently dosed, is providing insufficient analgesia so you order Hydromorphone 2 mg every 30-minutes as needed with up to 4 doses. Shortly thereafter x-rays reveal a closed, minimally displaced femur fracture and Orthopedic surgery promptly arrives to reduce the injury.

Ten minutes post-reduction, frantic family members scream for help and you dash into the room to find a blue patient with bradycardia. Fortunately, you quickly resuscitate your patient with a single dose of Narcan and less than one-minute of bag-valve mask ventilations plus chest compressions. Frustrated, you contemplate the fine line between adequate analgesia (patient-satisfaction) and iatrogenic injury (patient-safety). Wondering whether any validated tools to identify patients at high-risk for apnea exist, you turn to the medical literature.

PICO Question

Population: Adult patients requiring analgesia or sedation

Intervention: Application of an apnea screening tool

Comparison: Apnea risk-stratification by clinical gestalt alone

Outcome: Prognostic accuracy, reliability, physician/patient acceptability, external validity.


First years: Likelihood ratios for a sleep apnea clinical prediction rule, Am J Respir Crit Care Med 1994; 150: 1279-1285. (http://pmid.us/7952553)

Second years: STOP Questionnaire: A tool to screen patients for obstructive sleep apnea. Anesthesiology 2008; 108: 812-821. (http://pmid.us/18431116)

Third years: Validation of the Berlin Questionnaire and American Society of Anesthesiologists Checklist as screening tools for obstructive sleep apnea in surgical patients , Anesthesiology 2008; 108: 822-830. (http://pmid.us/18431117)

Fourth years: Diagnosis and initial management of obstructive sleep apnea without polysomnography, Annals Intern Med 2007; 146: 157-166. (http://pmid.us/17283346).


Article 1: Likelihood Ratios for a Sleep Apnea Clinical Prediction Rule Am J Respir Crit Care Med 1994;150:1279-1285

Article 2: STOP Questionnaire – A Tool to Screen Patients for Obstructive Sleep Apnea Anesthesiology 2008; 108:812-8

Article 3: Validation of the Berlin Questionnaire and American Society of Anesthesiologists Checklist as Screening Tools for Obstructive Sleep Apnea in Surgical Patients, Anesthesiology 2008;108:822-830

Article 4: Diagnosis and Initial Management of Obstructive Sleep Apnea without Polysomnography – A Randomized Validation Study Annals of Internal Medicine 2007; 146: 157-166

Bottom Line

For EM purposes, the STOP-Bang questionnaire appears superior to the Berlin Questionnaire and the ASA screen for the rapid identification of OSA. In one populations’ pre-op patients, the STOP-Bang questionnaire reliably identified a low risk subset of patients who are unlikely to have AHI-defined OSA. Before widespread application of this clinical decision rule, external validation should confirm the diagnostic accuracy of STOP-Bang in other patient populations. Contrary to classic teaching, high-risk OSA patients can immediately begin CPAP therapy following brief screening in lieu of definitive polysomnography (PSG) without impairing 3-month sleep apnea events, quality of life or adverse events. In fact, the non-PSG approach appears to improve compliance and is favored by patients. The authors’ offer a feasible alternative to PSG in the form of risk-stratification with ESS, SACS, and RDI with 94% pre-test probability of OSA. If validated in ED populations, such an approach could identify high-risk OSA patients allowing immediate institution of CPAP while waiting criterion standard PSG testing. In already over-crowded ED’s such interventions would need to demonstrate reduced short-term morbidity and not compromise the care of other ED patients.