Prediction of difficult mask ventilation

Background: Maintenance of ah-way patency and oxygenation are the main obje ctives of face-mask ventilation. Because the incidence of difficult mask ve ntilation (DMV) and the factors associated with it are nor well known, we u ndertook this prospective study. Methods: Difficult mask ventilation was defined as the inability of an unas sisted anesthesiologist to maintain the measured oxygen saturation as measu red by pulse oximetry > 92% or to prevent or reverse signs of inadequate ve ntilation during positive-pressure mask ventilation under general anesthesi a. A univariate analysis was performed to identify potential factors predic ting DMV, followed by a multivariate analysis, and odds ratio and 95% confi dence interval were calculated. Results: A total of 1,502 patients were prospectively included. DMV was rep orted in 75 patients (5%; 95% confidence interval, 3.9-6.1%), with one case of impossible ventilation. DMV was anticipated by the anesthesiologist in only 13 patients (17% of the DMV cases). Body mass index, age, macroglossia , beard, lack of teeth, history of snoring, increased Mallampati grade, and lower thyromental distance were identified in the univariate analysis as p otential DMV risk factors. Using a multivariate analysis, five criteria wer e recognized as independent factors for a DMV (age older than 55 yr, body m ass index > 26 kg/m(2), beard, lack of teeth, history of snoring), the pres ence of two indicating high likelihood of DMV (sensitivity, 0.72; specifici ty, 0.73). Conclusion: In a general adult population, DMV was reported in 5% of the pa tients. A simple DMV risk score was established. Being able to more accurat ely predict DMV may improve the safety of airway management.