A novel method to detect accidental oesophageal intubation based on ventilation pressure waveforms

Abstract

Background: Emergency endotracheal intubation results in accidental oesophageal intubation in up to 17% of patients. This is frequently undetected thereby adding to the morbidity and mortality. No current method to detect accidental oesophageal intubation in an emergency setting is both highly sensitive and specific. We hypothesized that, based on differences between the mechanical properties of the oesophagus and the trachea/lung, ventilation pressures could discriminate between tracheal and oesophageal intubation. Such a technique would potentially not suffer some of the limitations of current methods to detect oesophageal intubation in emergency conditions such as noisy environment (making clinical assessment difficult) or low/no flow states (reducing the applicability of capnometry). The aim of our study was thus to develop and assess a technique that may more rapidly and accurately differentiate oesophageal from tracheal intubation based on airway pressure gradients.

Materials and methods: Forty adult patients undergoing elective surgery were included. In 20 patients the trachea was intubated with an endotracheal tube; in 20 patients the oesophagus was purposefully intubated using an Easytube(®) (Rüsh, Germany). In all patients, a thin air-filled catheter was inserted through the tube lumen until its tip was 1cm from the distal end, and connected to a pressure transducer. Pressure was recorded simultaneously from a second catheter at the proximal end of the tube. For the first three manual ventilations in each patient, a parameter (D) based on temporal (dP/dt) and spatial (dP/ds) pressure gradients (and reflecting flow divided by elastance) was calculated and evaluated for its ability to discriminate between oesophageal and tracheal intubation.

Results and discussion: For all tracheal ventilations, D-values were >0.5 (range 0.6-47.9), while for all oesophageal ventilations D-values were <0.5 (range 0.0005-0.07).

Conclusion: This technique has the potential to provide a diagnosis of failed intubation within seconds with high sensitivity and specificity.

Trial registration: ClinicalTrials.gov NCT01080508.