Temporal comparison of ultrasound vs. auscultation and capnography in verification of endotracheal tube placement

Acta Anaesthesiol Scand 2011; 55: 1190–1195 Printed in Singapore. All rights reserved © 2011 The Authors Acta Anaesthesiologica Scandinavica © 2011 The Acta Anaesthesiologica Scandinavica Foundation ACTA ANAESTHESIOLOGICA SCANDINAVICA doi: 10.1111/j.1399-6576.2011.02501.x Temporal comparison of ultrasound vs. auscultation and capnography in verification of endotracheal tube placement P. Pfeiffer1, S. S. Rudolph2, J. Børglum2 and D. L. Isbye2 1 Akutcentrum/Anestesikliniken, Skåne University Hospital, SUS Malmø, Sweden and 2Department of Anaesthesia and Intensive Care Medicine, Copenhagen University Hospital, Bispebjerg, Denmark Background: This study compared the time consumption of bilateral lung ultrasound with auscultation and capnography for verifying endotracheal intubation. Methods: A prospective, paired, and investigator-blinded study carried out in the operating theatre. Twenty-five adult patients requiring endotracheal intubation were included. During intubation, transtracheal ultrasound was performed to visualize passage of the endotracheal tube. During bag ventilation, bilateral lung ultrasound was performed for the detection of lung sliding as a sign of ventilation simultaneous with capnography and auscultation of the epigastrium and chest. Primary outcome measure was time difference to confirmed endotracheal intubation between ultrasound and auscultation alone. Secondary outcome measure was time difference between ultrasound and auscultation combined with capnography. Results: Both methods verified endotracheal tube placement in all patients. In 68% of patients, endotracheal tube placement was visualized by real-time transtracheal ultrasound. Compar- ing ultrasound with the combination of auscultation and capnography, there was a significant difference between the two methods. Median time for ultrasound was 40 s [interquartile range (IQR) 35–48 s] vs. 48 s (IQR 45–53 s), P < 0.0001. Mean difference was -7.1 s in favour of ultrasound [95% confidence interval (CI) -9.4–-4.8 s]. No significant difference was found between ultrasound compared with auscultation alone. Median time for auscultation alone was 42 s (IQR 37–47 s), P = 0.6, with a mean difference of -0.88 s in favour of ultrasound (95% CI -4.2– 2.5 s). Conclusions: Verification of endotracheal tube placement with ultrasound is as fast as auscultation alone and faster than the standard method of auscultation and capnography. S There is an increasing focus on the utility of ultrasound in airway management and accordingly, The American College of Emergency Physicians recently published a policy statement on the verification of endotracheal intubation mentioning ultrasound as a possible future adjunct.6 Ultrasound can confirm endotracheal intubation in a number of ways. By scanning at the level of the cricoid membrane, direct visualization of the passage of the endotracheal tube through the vocal chords is possible.7,8 By scanning of the anterior neck at the level of the suprasternal notch, the absence of oesophageal shadowing in the paratracheal tissue following intubation indicate endotracheal tube placement.9 Real-time visualization of endotracheal tube placement may be possible ecuring the airway by endotracheal intubation is a fundamental skill in anaesthesia, critical care, and emergency medicine. Oesophageal or main stem intubation are well-known complications to this procedure which, if unrecognized, can lead to increased morbidity and mortality.1–3 Numerous methods are used for verifying correct placement of the endotracheal tube, but no single method has been shown to be fail-safe, and therefore, combinations of different methods are used in clinical practice.4,5 One such combination commonly used for confirming endotracheal intubation is clinical evaluation by auscultation of the epigastrium and chest supplemented with measurement of end-tidal CO2 (ETCO2) by capnography or capnometry. 1190 Accepted for publication 26 June 2011 © 2011 The Authors Acta Anaesthesiologica Scandinavica © 2011 The Acta Anaesthesiologica Scandinavica Foundation Lung ultrasound for ET tube verification by transtracheal scanning just below the cricoid ring.9,10 Finally, during post-intubation, bag ventilation diaphragmatic excursions visualized by two-dimensional ultrasound or the identification of specific dynamic ultrasonographic signs in the pleural interface known as lung sliding has been shown to be both sensitive and specific for lung ventilation and thus endotracheal tube position.8,11–14 In this context, the absence of lung sliding over one lung may indicate main stem intubation or pneumothorax, and absence over both lungs could indicate either oesophageal intubation11 or cuff leakage. However, no studies have compared the temporal relationship between the use of ultrasound for detection of lungsliding vs. standard clinical verification by auscultation alone or in combination with capnography for endotracheal intubation. The goal of this study was to compare the use of bilateral lung ultrasound to the clinical standard of auscultation and capnography for confirming endotracheal intubation. The primary outcome measure was time difference to confirmed endotracheal intubation between ultrasound and auscultation alone. The secondary outcome measure was time difference to confirmed endotracheal intubation between ultrasound and auscultation combined with capnography. Methods This study was designed as a prospective, paired, and investigator-blinded study and carried out during daytime in March 2010 in the operating suite of an urban teaching hospital in Copenhagen, Denmark. The Committees on Biomedical Research Ethics for the Capital Region of Denmark approved the study (journal number H3-2009-105), and written informed consent was obtained from all patients. This study was not registered as a clinical trial as the interventions in this study had no direct influence on the patients’ care, and participation in the study carried no risk for the patient. Patients eligible for study enrolment were 18 years and older who were admitted for elective or acute surgery and planned for endotracheal intubation. Exclusion criteria were history of previous difficult intubation or suspected difficult intubation due to abnormal airway anatomy. The physician responsible for all ultrasound investigations had a total of 3 hours of workshops and didactic lessons on lung ultrasound before conducting the study. Anaesthesia was induced with an opioid, propofol, or thiopental, and neuromuscular blockade was performed with succinylcholine 1 mg/kg. The trachea was intubated via direct laryngoscopy with endotracheal tubes sized 7.0 mm for women and 8.0 mm for men, and the use of a stylet was at the discretion of the anaesthesiologist responsible for the intubation. Three investigators were involved in the practical conduct of the study: one anaesthesiologist performing the intubation and subsequent bag ventilation of the patient, one anaesthesiologist performing the ultrasound, and one anaesthesiologist administering drugs and performing the auscultation. The investigators performing auscultation and ultrasound were blinded from each other by a drape hung across the patient’s chest (Fig. 1). Furthermore, the ultrasonographer was blinded from the verbal communication between the other investigators by wearing headphones playing white noise. Ultrasound investigation was performed with a Sonosite S-ICU equipped with a HFL 38x linear array 13–6 MHz transducer (Sonosite, Bothell, WA, USA). To determine optimal transducer position for visualization of the pleura, a two-dimensional realtime ultrasound scan was performed during preoxygenation with the transducer placed in the second or third intercostal space in the midclavicular line bilaterally. During intubation, the transducer was placed in the suprasternal notch for transtracheal real-time visualization of the passage of the endotracheal tube. Following intubation and prior to bag ventila- Fig. 1. Photo showing study set-up. Post-intubation bag ventilation is performed, and both investigators indicate that lung ventilation has been confirmed. 1191 P. Pfeiffer et al. tion, the transducer was moved to the right side of the chest and subsequently to the left side for detection of lung sliding as indication of bilateral lung ventilation. Standard protocol was followed for auscultation with the investigator first auscultating over the epigastrium, then in the right and left lung in that order. Unchanged ETCO2 levels and capnography after six ventilations were regarded as final proof of endotracheal intubation.4,15,16 Age, height, body mass index (BMI), American Society of Anesthesiologists (ASA) physical status score, and Cormack grading of direct laryngoscopy were recorded. All intubations were recorded on camera and stored as AVI files for later analysis. Furthermore, 60-s ultrasound clips of the intubation procedures were recorded. Time measurement started when the laryngoscope blade was introduced into the mouth. Ultrasonographic detection of lung sliding on both sides of the chest was indicated by inaudible signals to the camera, as were auscultation over the epigastrium and in both sides of the chest. Finally, the anaesthesiologist performing the ventilations indicated six bag ventilations with unchanged capnography. Afterwards, times to the different end points were measured by reviewing the recorded clips. For statistical analysis, the SAS system ver. 8.2 (SAS institute inc., Cary, NC, USA) was used. Age was reported as median with range. BMI, ASA score, and absolute time were reported as median with 25% to 75% range [interquartile range (IQR)]. Difference in time was calculated as ‘new method (ultrasound) minus old method’ and reported as mean with 95% confidence interval (CI). For comparison of methods, a paired t-test was used, and P-values less than 0.05 were considered statistically significant. Sample size calculation In a small pilot study on 10 patients, the standard deviation (SD) of time to verification of endotracheal tube placement by auscultation was approximately 10 s. It was estimated that the SD of the difference in time between the two methods would be approximately 5 s. We aimed at recognizing a difference of 1 SD, as a difference of 5 s in practice would mean that the two methods were equal. Using a significance level of 0.05, a power of 90% could be reached by enrolling 21 patients in the study, but it was decided that 25 patients will be enrolled to allow for dropouts. 1192 Results In total, 25 patients were enrolled in the study, 12 men and 13 women. Median age was 49 years (range 22–88 years). Median BMI was 25 (IQR 24–27.5), and median ASA physical status score was 1 (IQR 1–2). All 25 patients were intubated by direct laryngoscopy. Twenty-three patients were classified as Cormack grade 1, and two were classified as Cormack grade 2. In three patients, rapid sequence induction was applied according to department guidelines. No oesophageal intubations occurred, and none of the endotracheal tubes needed repositioning during anaesthesia. In 17 patients (68%), passage of the endotracheal tube through the trachea was visualized in real time by ultrasound. Comparing ultrasound with auscultation alone, we found no significant difference in time to final verification between the two methods. Median time for verification by ultrasound was 40 s (IQR 35–48 s) vs. 42 s (IQR 37–47 s) for auscultation alone, P = 0.6, with a mean difference of -0.88 s in favour of ultrasound (95% CI -4.2–2.5 s). Comparing ultrasound with the combination of auscultation and capnography, there was a significant difference between the two methods. Median time to verification by auscultation and six uniform curves on the capnograph was 48 s (IQR 45–53 s), P < 0.0001. Mean difference was -7.1 s in favour of ultrasound (95% CI -9.4–-4.8 s). Both methods verified endotracheal tube placement in all patients. In one patient, more than six ventilations were required to verify endotracheal intubation by auscultation, and in two others, more than six ventilations were required to verify it by ultrasound. Median time for the administration of the six ventilations was 14 s (IQR 13–17), corresponding to a ventilatory rate of approximately 26 breaths per minute. In six patients, auscultation alone was faster than ultrasound, and in seven patients, ultrasound and auscultation alone were equally fast. Discussion The principal finding of this study is that the use of bilateral lung ultrasound for verification of endotracheal intubation is as fast as auscultation. Furthermore, we found that bilateral lung ultrasound was significantly faster than the combination of auscultation and ETCO2 measurement for six breaths. In addition, it is worth noticing that the ultrasonogra- Lung ultrasound for ET tube verification pher had limited experience with lung ultrasound, but performed as well as the standard method in confirming endotracheal intubation. Finally, we were able to visualize real-time tube passage in only 68% of the intubations. We therefore believe that transtracheal ultrasonography for visualization of the tube passage at this level cannot be relied on as an indication of endotracheal tube placement alone. Two recent studies have reported the use of ultrasound in the verification of endotracheal intubation. In a study of 30 emergency department intubations, the combination of transcricothyroid membrane ultrasound and lung sliding for verification of endotracheal intubation was described.8 With high sensitivity and specificity, the authors found this to be a useful technique for confirming endotracheal intubation in the emergency department, even though they also argued that the ultrasound method was time consuming. Another study investigated the use of a brief ultrasound examination of the chest for lung sliding and diaphragmatic movements to confirm correct placement of double-lumen tubes.17 The results indicated that ultrasound added to clinical assessment ensured more precise placement of left-sided double-lumen tubes than did clinical assessment alone. By selective clamping of the bronchial and tracheal limbs with subsequent ultrasound scan for lung sliding, the investigators were able to distinguish between ventilation and absence of ventilation on both sides of the chest. We included patients undergoing both elective and acute surgery, but none of our patients needed urgent airway management. Comparing the use of ultrasound scanning on the anterior chest with combined auscultation and capnography, we found ultrasound to be at least as fast as the conventional method. Contrary to previous studies, we did not find the ultrasound method to be time consuming, but trained staff with unhindered access to portable ultrasound machines with short start-up times is essential in limiting time consumption.13 We “blinded” the ultrasonographer from the verbal communication in the operating suite by wearing headphones playing white noise, simulating a noisy environment. Our results indicate that the ultrasonographer needs only vision – not hearing – to confirm endotracheal intubation and bilateral lung ventilation. Scanning on the anterior chest wall for lung sliding as we have done has been debated, particularly in the trauma patient where subcutaneous emphysema, haematomas, or haemothorax or pneu- mothorax may influence ultrasound imaging.8,14 For these reasons, a more lateral approach or the use of transcricothyroid ultrasound for verification of endotracheal intubation has been suggested. As shown in a previous study, lung sliding may be less reliable in differentiating tracheal from main stem intubation in cadavers, but we believe that postmortem changes in the lung tissue may be more outspoken with the more dependent lateral probe position used in that study.7 In our opinion, by performing a pre-intubation ultrasound scan of the anterior chest in the spontaneously breathing patient, it is possible to locate optimal probe position for the detection of lung sliding. With minimal loss of time, this can be performed either simultaneously with well-known standard ultrasound protocols to detect or exclude significant thoracic pathology, e.g. an anterior pneumothorax18–20 or during pre-oxygenation. No single method for verification of endotracheal intubation is failsafe. Direct visualization of tube passage and subsequent position between the vocal cords should always be attempted, but this may be impossible due to factors such as patient anatomy, positioning, or airway pathology. Auscultation for verification of endotracheal intubation may be impractical in noisy environments or may be difficult to interpret due to pathological breath sounds. For ETCO2 measurement, values can be misleading in patients with low flow states or cardiac arrest, and main stem intubation may not be recognized by ETCO2 measurement alone until later in the postintubation period.4,5,11,14 In case of oesophageal intubation, ingestion of antacids or carbonated drinks and prolonged bag–mask ventilation prior to intubation can result in misleading ETCO2 measurement, and accordingly, a total of six capnography curves of normal configuration for confirming endotracheal intubation has been proposed.4,21 The clinical significance of this study stands out when considering the findings of the present study and the limitations of the traditional methods for verification of endotracheal intubation. As shown in previous studies, lung ultrasound is highly specific and sensitive for verification of endotracheal intubation,8,14,22 and the present data indicate that ultrasound for verification of endotracheal intubation by lung sliding is equally fast as auscultation and faster than the combination of auscultation and ETCO2 detection. With the use of ultrasound becoming more widespread in a multitude of clinical settings and the technology becoming both cheaper and increasingly portable, lung ultrasound may have a 1193 P. Pfeiffer et al. future role as an adjunct for verification of endotracheal intubation, especially in selected settings such as the pre-hospital environment, during interhospital transfer or aeromedical retrieval of intubated patients. With a protocolized approach to using the ultrasound machine in the intubation procedure, time spent for preparation should be minimal. Switching on and setting up the machine can be performed well in advance of the intubation, and preparing the patient for the pre-scan by exposing the upper torso is no different than preparing him for auscultation. The necessary skill level is easily obtainable as observed in the present study. There are several limitations to this study. First, this was a single-centre study with a small sample size in a controlled environment. To minimize interinvestigator variation in this small sample size, all interventions were performed by the same investigator throughout the study, and this may overestimate the accuracy of the results, particularly with the ultrasonographer gaining experience as the study went on. Furthermore, we detected no main stem intubations but did not exclude main stem intubation by chest X-ray or fibre-optic inspection. These were omitted as auscultation in combination with capnography is considered the practical standard to verify tracheal placement of the tube. Thus, we cannot report any results concerning difference in the time consumption between ultrasound and the conventional methods in detecting oesophageal and endobronchial intubations. Investigator bias was sought minimized by visual as well as auditory blinding between the ultrasonographer and the auscultating investigator. None of these investigators were blinded from the intubating anaesthesiologist as this, in our opinion, would be highly unlikely under normal circumstances. The intubating anaesthesiologist was responsible for the bag ventilation required for ETCO2 measurement, and investigator bias is possible in this regard by wilfully delaying ventilations. However, our median time for administration of the six ventilations corresponds to a supranormal respiratory rate of 26 breaths per minute, which should minimize this aspect of investigator bias. Further studies should evaluate the use of ultrasound for verification of endotracheal intubation in various patient categories and settings such as the obese patients or the patients with predicted difficult airway, as well as utility in the acute/trauma setting or in the out-of-hospital environment. 1194 Conclusion We conclude that verification of endotracheal tube placement by bilateral lung ultrasound is as fast as auscultation alone and faster than the standard method of auscultation and capnography. Conflict of interest: The authors have no conflicting interests to declare. The study was conducted with departmental funding only. References 1. Schwartz DE, Matthay MA, Cohen NH. Death and other complications of emergency airway management in critically ill adults. A prospective investigation of 297 tracheal intubations. Anesthesiology 1995; 82: 367–76. 2. Adnet F, Jouriles NJ, Le Toumelin P, Hennequin B, Taillandier C, Rayeh F, Couvreur J, Nougiere B, Nadiras P, Ladka A, Fleury M. Survey of out-of-hospital emergency intubations in the French Prehospital Medical System: a multicenter study. Ann Emerg Med 1998; 32: 454–60. 3. Miller JS, Itani KM, Oza MD, Wall MJ. Gastric rupture with tension pneumoperitoneum: a complication of difficult endotracheal intubation. Ann Emerg Med 1997; 30: 343–6. 4. Grmec S. Comparison of three different methods to confirm tracheal tube placement in emergency intubation. Intensive Care Med 2002; 28: 701–4. 5. Knapp S, Kofler J, Stoiser B, Thalhammer F, Burgmann H, Posch M, Hofbauer R, Stanzel M, Frass M. The assessment of four different methods to verify tracheal tube placement in the critical care setting. Anesth Analg 1999; 88: 766–70. 6. American College of Emergency Physicians. Clinical Policy Committee. Verification of endotracheal tube placement. Ann Emerg Med 2009; 54: 141–2. 7. Ma G, Davis DP, Schmitt J, Vilke GM, Chan TC, Hayden SR. The sensitivity and specificity of transcricothyroid ultrasonography to confirm endotracheal tube placement in a cadaver model. J Emerg Med 2007; 32: 405–7. 8. Park SC, Ryu JH, Yeom SR, Jeong JW, Cho SJ. Confirmation of endotracheal intubation by combined ultrasonographic methods in the emergency department. Emerg Med Australas 2009; 21: 293–7. 9. Werner SL, Smith CE, Goldstein JR, Jones RA, Cydulka RK. Pilot study to evaluate the accuracy of ultrasonography in confirming endotracheal tube placement. Ann Emerg Med 2007; 49: 75–80. 10. Drescher MJ, Conard FU, Schamban NE. Identification and description of esophageal intubation using ultrasound. Acad Emerg Med 2000; 7: 722–5. 11. Chun R, Kirkpatrick AW, Sirois M, Sargasyn AE, Melton S, Hamilton DR, Dulchavsky S. Where’s the tube? Evaluation of hand-held ultrasound in confirming endotracheal tube placement. Prehosp Disaster Med 2004; 19: 366–9. 12. Hsieh KS, Lee CL, Lin CC, Huang TC, Weng KP, Lu W. Secondary confirmation of endotracheal tube position by ultrasound image. Crit Care Med 2004; 32: S374–S7. 13. Sustic A. Role of ultrasound in the airway management of critically ill patients. Crit Care Med 2007; 35: S173–7. 14. Weaver B, Lyon M, Blaivas M. Confirmation of endotracheal tube placement after intubation using the ultrasound sliding lung sign. Acad Emerg Med 2006; 13: 239–44. 15. Nagler J, Krauss B. Capnography: a valuable tool for airway management. Emerg Med Clin North Am 2008; 26: 881–97. 16. Salem MR. Verification of endotracheal tube position. Anesthesiol Clin North America 2001; 19: 813–39. Lung ultrasound for ET tube verification 17. Sustić A, Protić A, Cicvarić T, Zupan Z. The addition of a brief ultrasound examination to clinical assessment increases the ability to confirm placement of double-lumen endotracheal tubes. J Clin Anesth 2010; 22: 246–9. 18. Lichtenstein D. Lung ultrasound in acute respiratory failure: an introduction to the BLUE-protocol. Minerva Anestesiol 2009; 75: 313–7. 19. Lichtenstein DA, Menu YA. Bedside ultrasound sign ruling out pneumothorax in the critically ill. Lung sliding. Chest 1995; 108: 1345–8. 20. Rippey JC, Royse AG. Ultrasound in trauma. Best Pract Res Clin Anaesthesiol 2009; 23: 343–62. 21. Linko K, Paloheimo M, Tammisto T. Capnography for detection of accidental oesophageal intubation. Acta Anaesthesiol Scand 1983; 27: 199–202. 22. Milling TJ, Jones M, Khan T. Transtracheal 2-D ultrasound for identification of esophageal intubation. J Emerg Med 2007; 32: 409–14. Address: Peter Pfeiffer Dalgas Boulevard 134 4th DK – 2000 F Denmark e-mail: ptrpfffr@gmail.com 1195