Confirmation of endotracheal tube placement after intubation using the ultrasound sliding lung sign.

OBJECTIVES: To evaluate the performance of the ultrasound (US) sliding lung sign as a predictor of endotracheal tube (ETT) placement. Many other tools and examination findings have been used to confirm ETT placement; erroneous placement of the ETT has even been confirmed by US.
METHODS: This was a laboratory study using fresh, recently dead cadavers. Cadavers were obtained at a medical school anatomy laboratory on the basis of availability during a four-month period. Subjects who died from significant trauma or after thoracic surgery were excluded. A numerical randomization tool was used to direct where the tube would be placed on intubation. Laryngoscopy was performed, and the ETT was placed in the esophagus, in the trachea, or in the right main stem (RMS) bronchus. Placement was confirmed by direct laryngoscopic visualization of ETT passage through vocal cords or with fiber optic visualization, as needed. US images of the sliding lung sign, sliding of visceral and parietal pleura past each other, were taken on both sides of the chest at the mid axillary line during ventilation with an ambu bag. Two board-certified emergency physicians with hospital credentialing in emergency US used a 4-2 MHz micro-convex transducer on a Sonosite 180 Plus for imaging. The sonologists were blinded to the location of the endotracheal tube and imaged and recorded their results individually. A positive sliding lung sign was taken to signify lung expansion with ventilation in a hemithorax. Endotracheal versus esophageal ETT placement, as well as tracheal versus RMS, was determined on the basis of sliding lung findings on both sides of the chest. Interpreter agreement, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratios (LHR) were calculated for tracheal (including RMS) versus esophageal, as well as main trachea versus RMS intubation.
RESULTS: Nine cadavers yielded 68 intubations. For esophageal versus tracheal (including RMS) intubation, sonologist 1 (S1) had a sensitivity of 95.4% (95% CI = 84.2% to 99.4%), a specificity of 100% (95% CI = 86.3% to 100%), an NPV of 92.6% (95% CI = 75.7% to 99.1%), and a PPV of 100% (95% CI = 91.4% to 100%) with an LHR of 0.05 (95% CI = 0.01 to 0.2) for a negative test. Sonologist 2 (S2) had a sensitivity of 100% (95% CI = 91.8% to 100%), a specificity of 100% (95% CI = 86.3% to 100%), an NPV of 100% (95% CI = 86.3% to 100%), and a PPV of 100% (95% CI = 91.8% to 100%); agreement was 97% (kappa = 0.94; 95% CI = 0.7 to 1.2). In RMS versus tracheal, S1 had a sensitivity of 69.2% (95% CI = 48.2% to 85.7%), a specificity of 93.3% (95% CI = 68.1% to 99.8%), a PPV of 94.7% (95% CI = 73.9% to 99.9%), and an NPV of 63.6% (95% CI = 40.7% to 82.8%) with an LHR for a positive test of 10.4 (95% CI = 2.2 to 59.1) and of 0.4 (95% CI = 0.2 to 0.6) for negative test. S2 had a sensitivity of 78.6% (95% CI = 59.1% to 91.7%), a specificity of 100% (95% CI = 78.2% to 100%), a PPV of 100% (95% CI = 84.6% to 100%), NPV of 71.4% (95% CI = 47.8% to 88.7%), with an LHR for a negative test of 0.2 (95% CI = 0.1 to 0.4); agreement was 85.9% (kappa = 0.6; 95% CI = 0.4 to 0.9).
CONCLUSIONS: These results show that US imaging of the sliding lung sign in a cadaver model is an accurate method for confirmation of ETT placement. Further, the technique may have some utility in differentiating RMS bronchus from main tracheal intubations.