REAL-TIME ULTRASOUND GUIDED ENDOTRACHEAL INTUBATION
Description
Background End tidal carbon dioxide (EtCO2) waveform (capnography) is considered to be the gold standard in confirming the position of the endotracheal tube. Real-time Ultrasonography is a new method to confirm endotracheal intubation in situations where capnography is not available. Methods During intubation the linear probe of ultrasound was placed over the cricothyroid membrane. Ultrasonographic images of endotracheal intubation shows ‘empty esophagus sign’ while esophageal intubation shows a ‘double trachea sign’. Simultaneously, another operator will connect the capnography and will observe the EtCO2 wave. Endotracheal intubation will be confirmed when square waveform is maintained with end-tidal carbondioxide greater than 4mm Hg. The accuracy, sensitivity, specificity and time taken to identify is noted using both the methods. Results Real-time Ultrasound was 98.46% ( 95% CI : 91.72% - 99.96%) accurate in confirming endotracheal intubation. The empty esophagus sign was 98.36% sensitive and 100% specific. Endotracheal intubation was confirmed earlier with real-time ultrasound (12.43±5.4sec) than capnography (16.68±5.50sec) and difference in time taken to confirm the position of the endotracheal tube was statistically significant with p<0.05. Conclusion Real-time tracheal ultrasound is an effective alternate method which is as sensitive and accurate as gold standard capnography to confirm endotracheal intubation. It can be used in clinical settings where capnography cannot be used as in massive pulmonary embolus and cardiac arrest or in settings where it is not available.
Files
Steps to reproduce
65 patients in the critical care unit above the age of 18 years requiring elective intubation were included. Patients with full stomach, pregnant women, pre-existing cardiopulmonary pathology, pre-existing laryngeal or tracheal pathology, previous neck surgeries/ radiation, anticipated difficult airway, patients requiring emergency intubation were excluded from the study. It was a cross sectional, observational study conducted in a tertiary care centre. The patients were induced with intravenous Injection Propofol 2mg/kg or Injection Etomidate 0.3mg/Kg depending on hemodynamics and paralysed with Injection Suxamethonium 1mg/kg iv. The investigator placed linear probe, 15Hz of ultrasound machine (GE Vivid TM) transversely over the suprasternal notch during direct laryngoscopy to identify the tracheal rings and finally the probe was moved cephalad and placed over the cricothyroid membrane to identify the vocal cords. Vocal cords were identified as paired hyper echoic linear structures. Ultrasonographic images of endotracheal intubation shows ‘empty esophagus sign’ while esophageal intubation shows a ‘double trachea sign’. Simultaneously, another operator connected the capnography sampling line to the monitor and observed the EtCO2 wave. Endotracheal intubation was confirmed when square waveform is maintained for five breaths. The time taken to assess the position of endotracheal tube was noted in both the methods with stopclock. T1 is defined from the time intubator is able to visualize the vocal cords till another operator confirms it simultaneously using the trans-tracheal ultrasonogram. Time taken from visualization of vocal cords till the appearance of five consecutive square wave pattern is taken as T2. Table 1 : Scoring system for ease of intubation INDEX YES NO Was introduction of laryngoscope difficult due to presence of USG probe? 1 0 Was Visualization of cords difficult? 1 0 Was the external laryngeal manipulation by probe not good? 1 0 Was introduction of the endotracheal tube difficult? 1 0 The minimum score will be 0. The maximum score will be 4. The ease of laryngoscopy will be graded depending on the cumulative score of all four indices. 0 – Easy intubation 1 - Minimal disturbance but no alteration in probe position, 2 - Minimal disturbance needed slight alteration in probe position, 3 - Disturbance needing major alteration in probe position, 4 - Profound difficulty. Sample size was calculated using Bland-Altman plot. A power of 90% and Alpha error 0.05 showed that a sample size of 62 would be required for the study. We conducted the study with a sample size of 65. Descriptive statistics were done for all data and were reported in terms of mean values and percentages. Suitable statistical tests of comparison were applied. Continuous variables were analyzed with the paired t test. Statistical significance was taken as p < 0.001. The data was analyzed using SPSS version 25 and Microsoft Excel 2010.