Detection of risk factors for difficult tracheal intubation

Several studies have identified difficult airway management including a difficult tracheal intubation of patients undergoing general anaesthesia as a major cause of anaesthesia-related morbidity and mortality. Therefore it is presumed that a difficult tracheal intubation is a surrogate marker for morbidity and mortality, and by reducing the prevalence of difficult tracheal intubation then morbidity and mortality will be reduced as well. From the Danish Anaesthesia Database (DAD), we retrieved a cohort of consecutive patients planned and attempted for tracheal intubation by direct laryngoscopy. Based upon various data including an intubation score registered in the database, we aimed to evaluate four different parameters, 'Obesity', 'avoidance of neuromuscular blocking agents', 'a previous difficult tracheal intubation' and 'the modified Mallampati-score', as possible risk factors for a difficult tracheal intubation. All of these risk factors were statistically associated with a difficult tracheal intubation, but the clinical significance varied substantially. However, neither 'obesity', 'the modified Mallampati-score' nor 'a previous difficult tracheal intubation' were sufficient as stand-alone tests for prediction of difficult tracheal intubation. In multivariate analyses the impact of obesity on the risk of difficult tracheal intubation seems weak, while both 'the modified Mallampati-score' and 'a previous difficult tracheal intubation' demonstrated to be clinically strong risk factors for difficult tracheal intubation. The evaluation of 'avoidance of neuromuscular blocking agents' as a risk factor differ substantially from the other assessments, as it concerns the impact of an intervention rather than of a patient-related risk factor for difficult tracheal intubation. In our assessment, 'avoiding neuromuscular blocking agents' was demonstrated as a possible risk factor for difficult and abandoned tracheal intubation independent of other risk factors recorded in the DAD. Several previous studies have failed to present specific risk factors that could identify difficult intubation or laryngoscopy by itself. Therefore it seems rational to focus on the development, testing and modification of multivariate models from large scale cohort studies, hereby making the prognostication operational in everyday clinical practice. From there the challenge may be to test the effectiveness of the use of such a model in order to evaluate whether it actually has the capability to reduce difficult tracheal intubation, complications, and mortality. It seems that such a trial should and could be conducted as a cluster randomized trial of anaesthesia departments within the framework of the DAD.