Comparison of two methods for predicting difficult intubation in obstetric patients

Preoperative airway assessment for prediction of difficult laryngoscopy and intubation was done using the modified Mallampati test and Wilson risk sum score in three hundred and seventy two obstetric patients undergoing elective as well as emergency Cesarean section under general anesthesia. 25 (6.7%) patients had laryngoscopy grade III or IV of whom 24 (6.4%) patients were difficult at tracheal intubation. Mallampati class III or IV predicted 15 of the 23 patients while Wilson risk sum score > or = 2 predicted 9 of the 14 patients in whom tracheal intubation was difficult. As a screening test for prediction of difficult intubation Wilson risk sum score was less sensitive (36%) but had almost same specificity (98.5%) and positive predictive value (64%) in comparison to modified Mallampati test (62.5%, 97.7% and 65% respectively). When used as a predictor of difficult laryngoscopy sensitivity, specificity and positive predictive value for modified Mallampati test were 60%, 97.6% and 65% respectively and for Wilson risk sum they were 36%, 98.5% and 64% respectively, but when both tests were combined as predictors (with either of tests positive) sensitivity improved to 100% while specificity was marginally decreased to 96.2% and positive predictive value (64.8%) remained almost the same. There was no significant association between age and laryngoscopy grade III or IV but there was significant (P < 0.01) relationship with weight and external laryngeal manipulation. The advantage of the above tests lies in, incorporating them into the preoperative protocol, rather than using them as sole predictors of difficult laryngoscopy and intubation.