Evaluation of predictors of a difficult laryngoscopy with an anaesthesia information management system (AIMS)

In the literature, difficult or impossible intubation is given with an incidence of 0.04 % to 13 %, depending on the definition and the patients investigated. The preoperative evaluation of the sterno- and thyreomental distance or of complex scores in order to predict a difficult intubation takes time. In addition to the Mallampati classification, we checked the influence of other medical parameters, which are recorded routinely during the preoperative consultation by anaesthesiologists, on the difficulty of the laryngoscopy (Cormack & Lehane Classification > II). For this, the data sets of more than 24,000 intubation anaesthesia procedures have been evaluated. In 1997 and 1998 the parameters gender, age, size, weight, dental status, status of nutrition, performing anaesthesiologist, ASA-, Mallampati-, Cormack & Lehane classification, degree of urgency, crush induction, and surgical department have been recorded using an anaesthesia information management system (AIMS). These data sets have been stored into a relational database during each anaesthesia procedure. For evaluation the information has been exported from the database to a statistics program via 'Structured Query Language' (SQL). In order to identify multivariate predictors for the occurrence of a difficult laryngoscopy, a gradual logistical regression analysis has been used. In addition to the modified Mallampati classification, the influence of the parameters gender, age, obesity (body mass index), dental status and ASA classification on the difficulty of a laryngoscopy could be proved. But only the Mallampati classification and obesity (body mass index) have been identified as predictors of a difficult laryngoscopy. The tested model of these two predictors has a specificity of 66 % and a sensitivity of 75 %. With the described parameters the anaesthetist has predictors for difficult intubation conditions at his disposal. These parameters are routinely recorded during the preoperative round by the anaesthesiologist. They can help to decide whether or not to use additional scores or whether or not to combine the Mallampati classification with further tests. Specificity and sensitivity of the model created hom these two parameters are rather low though. The recording of data using an AIMS has proved to be a suitable tool for the collection of data of importance for patients' safety.