Automatic detection of oesophageal intubation based on ventilation pressure waveforms shows high sensitivity and specificity in patients with pulmonary disease. (August 2016)

• Record Type: Journal Article
• Authors: Kalmar, Alain F.; Absalom, Anthony; Rombouts, Pieter; Roets, Jelle; Dewaele, Frank; Verdonck, Pascal; Stemerdink, Arjanne; Zijlstra, Jan G.; Monsieurs, Koenraad G.
• Is Part Of: Resuscitation. Volume 105(2016)
• Page Start: 36
• Page End: 40
• Publication Date: 2016-08
• Subjects: Oesophageal intubation -- Emergency intubation -- Pulmonary compliance -- Automatic detection algorithm; Resuscitation -- Periodicals; Resuscitation -- Periodicals; Réanimation -- Périodiques; Electronic journals; 616.025
• DOI: 10.1016/j.resuscitation.2016.05.010
• Languages: English
• ISSNs: 0300-9572

Abstract

Abstract: Background: Unrecognised endotracheal tube misplacement in emergency intubations has a reported incidence of up to 17%. Current detection methods have many limitations restricting their reliability and availability in these circumstances. There is therefore a clinical need for a device that is small enough to be practical in emergency situations and that can detect oesophageal intubation within seconds. In a first reported evaluation, we demonstrated an algorithm based on pressure waveform analysis, able to determine tube location with high reliability in healthy patients. The aim of this study was to validate the specificity of the algorithm in patients with abnormal pulmonary compliance, and to demonstrate the reliability of a newly developed small device that incorporates the technology. Materials and methods: Intubated patients with mild to moderate lung injury, admitted to intensive care were included in the study. The device was connected to the endotracheal tube, and three test ventilations were performed in each patient. All diagnostic data were recorded on PC for subsequent specificity/sensitivity analysis. Results and discussion: A total of 105 ventilations in 35 patients with lung injury were analysed. With the threshold D -value of 0.1, the system showed a 100% sensitivity and specificity to diagnose tube location. Conclusion: The algorithm retained its specificity in patients with decreased pulmonary compliance. We also demonstrated the feasibility to integrate sensors and diagnostic hardware in a small, portable hand-held device for convenient use in emergency situations.