Inter-observer reliability of ten tests used for predicting difficult tracheal intubation

To determine inter-observer reliability of ten preoperative airway assessment tests used for predicting difficult tracheal intubation. We prospectively assessed 59 patients undergoing elective surgery requiring tracheal intubation at a large metropolitan teaching hospital. Two experienced observers independently conducted the airway assessment tests on the same group of patients. Inter-observer reliability was examined using Kappa (K) and intraclass correlation coefficient (ICC). Two tests--mouth opening (ICC = 0.93) and chin protrusion (ICC = 0.89)--had excellent inter-observer reliability. Seven tests--thyromental distance (ICC - 0.74), subluxation (K = 0.66), atlanto-occipital extension distance (ICC = 0.67) and angle (K = 0.66), profile classification (K = 0.58), ramus length (ICC = 0.53), oropharyngeal best view (K = 0.49)--were moderately reliable. One test--Mallampati technique of assessing oropharyngeal view (K = 0.31)--had poor reliability. Many of the preoperative airway tests have only moderate inter-observer reliability. This may provide some insight into why previous research has failed to show that the tests accurately predict difficult tracheal intubation.