Derivation of the Difficult Airway Physiological Score (DAPS) in adults undergoing endotracheal intubation in the emergency department

Abstract

Abstract Background Prediction of serious outcomes among patients with physiological instability is crucial in airway management. In this study, we aim to develop a score to predict serious outcomes following intubation in critically ill adults with physiological instability by using clinical and laboratory parameters collected prior to intubation. Method This single-center analytical cross-sectional study was conducted in the Emergency Department from 2016 to 2020. The airway score was derived using the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) methodology. To gauge model’s performance, the train-test split technique was utilized. The discrete random number generation approach was used to divide the dataset into two groups: development (training) and validation (testing). The validation dataset’s instances were used to calculate the final score, and its validity was measured using ROC analysis and area under the curve (AUC). By computing the Youden’s J statistic using the metrics sensitivity, specificity, positive predictive value, and negative predictive value, the discriminating factor of the additive score was determined. Results The mean age of the 1021 patients who needed endotracheal intubations was 52.2 years (± 17.5), and 632 (62%) of them were male. In the development dataset, there were 527 (64.9%) physiologically difficult airways, 298 (36.7%) post-intubation hypotension, 124 (12%) cardiac arrest, 347 (42.7%) shock index > 0.9, and 456 [56.2%] instances of pH < 7.3. On the contrary, in the validation dataset, there were 143 (68.4%) physiologically difficult airways, 33 (15.8%) post-intubation hypotension, 41 (19.6%) cardiac arrest, 87 (41.6%) shock index > 0.9, and 121 (57.9%) had pH < 7.3, respectively. There were 12 variables in the difficult airway physiological score (DAPS), and a DAPS of 9 had an area under the curve of 0.857. The accuracy of DAPS was 77%, the sensitivity was 74%, the specificity was 83.3%, and the positive predictive value was 91%. Conclusion DAPS demonstrated strong discriminating ability for anticipating physiologically challenging airways. The proposed model may be helpful in the clinical setting for screening patients who are at high risk of deterioration.