European Journal of Cardiovascular Medicine

Background: Most studies for the prediction of difficult airways have been done in Caucasians, and the cut-offs set by them may not be applicable to the Indian population. The purpose of the study was to compare the validity of the upper lip bite test and the modified Mallampati test in predicting difficult intubation in the Indian population. Methodology:  A total of 279 patients scheduled for elective surgical procedures requiring endotracheal intubation were enrolled, and the upper lip bite test (ULBT) and modified Mallampati test (MMT) were assessed preoperatively, and Cormack and Lehane’s grading was recorded while performing direct laryngoscopy. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and accuracy were calculated for the upper lip bite test, the modified Mallampati test, and for both tests in parallel. Results:  The percentage of patients with a difficult direct laryngoscopy was 12.9%. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and accuracy for the upper lip bite test were 80.56% ,98.35%, 48.94, 0.2,87.88% ,97.15 % and 96.06%; the corresponding values for the modified Mallampati test were 55.56% ,72.43% ,2.015 ,0.614, 22.98% ,96.67% and 70.25%; and the corresponding values when both tests are used in parallel are 94.44% ,71.19% ,3.28 ,0.08 ,32.69% ,98.86 % and 74.19%. Conclusion:  Based on the findings of the current study, the upper lip bite test was found to be superior to the modified Mallampati test in all aspects of diagnostic test evaluation and can be used as a good predictor of a difficult airway when the test is positive, although a negative test doesn't rule out difficult intubation. When both tests were used in parallel, the sensitivity goes up to 94% from 80% with ULBT used alone. This enables it to be used as a screening test for difficult intubation to predict most cases, but the low positive and negative likelihood ratios suggest the use is restricted in its efficient utility in many scenarios.