Secondary Logo

Journal Logo

Airway management

Training in placement of the left-sided double-lumen tube among non-thoracic anaesthesiologists: intubation model simulator versus computer-based digital video disc, a randomised controlled trial

Campos, Javier H; Hallam, Ezra A; Ueda, Kenichi

Author Information
European Journal of Anaesthesiology: March 2011 - Volume 28 - Issue 3 - p 169-174
doi: 10.1097/EJA.0b013e328340c332



Recent advances in thoracic, cardiac, orthopaedic and oesophageal surgery have led to an increased need for lung separation techniques. In addition, the need to use lung isolation devices includes emergency rooms or surgical intensive care units where a potential for lung contamination exists and requires placement by an emergency or critical care physician. Accordingly, anaesthesiologists or other physicians with limited thoracic experience are increasingly called upon to place double-lumen endotracheal tubes (DLTs). Our studies have shown that anaesthesiologists with limited thoracic anaesthesia experience often fail to correctly place DLTs.1 Our recent study showed a high incidence (38%) of unrecognised malpositions or complete failure to place several lung isolation devices. Increased clinical experience would likely reduce this failure rate, but greater experience may not be possible, particularly for anaesthesiologists working in centres that perform relatively few thoracic cases. Therefore, improved non-clinical training methods are needed.

Anaesthesia simulators have been used to enhance learning and to improve performance,2 usually under the personal direction of an experienced clinician. Therefore, one educational approach to lung isolation techniques might involve training on an airway simulator mentored by an experienced thoracic anaesthesiologist. An alternate method of education could be the use of computer-based audiovisual systems, which have also been shown to improve performance, in some cases equal to that achieved with simulators.3,4 Such an educational method might involve the review of a digital video disc (DVD) containing all necessary didactic information (with video images) related to placement of DLTs.

The objective of this study was to compare the effectiveness of training with an airway model simulator versus DVD-based instruction. Successful and timely placement of the DLT by anaesthesiologists with limited experience in thoracic anaesthesia was evaluated.

Participants and methods

After approval by our Human Subjects Committee (Human Subjects Office, the University of Iowa, Iowa City, Iowa, USA), written informed consent was obtained from each of the 60 patients who participated in the study and also from each of the 27 participating anaesthesiologists. The anaesthesiologists were either faculty members (n = 23) or fellows (n = 4) in the Department of Anesthesia at the University of Iowa. Patients (age range 19–84 years; weight range 48–125 kg) undergoing elective thoracic or oesophageal surgery procedures for which one-lung ventilation (OLV) was required were included in the study. Anaesthesiologists were randomised in a parallel fashion in a 1: 1 ratio. Exclusion criteria for the patients included a history of difficult airway/intubation or a Mallampati class III or IV airway as determined during pre-operative evaluation. The inclusion criterion for the participating anaesthesiologists was to have familiarity with a DLT. Anaesthesiologists who had performed an anaesthetic procedure involving lung isolation more than twice in the preceding month were excluded.

Twenty-seven anaesthesiologists were randomly assigned to a training session with the airway simulator or a computer-based training DVD. After the training session, they tested their learning skills on consented patients undergoing surgical procedures that required lung isolation. Primary outcome recorded was successful placement of the DLT. Secondary outcomes included time taken for placement and the number of times the fibreoptic bronchoscope was used.

Airway simulator group

Anaesthesiologists assigned to the airway simulator group participated in a hands-on training session the afternoon prior to surgery. A thoracic anaesthesiologist (JHC or KU) demonstrated placement of a DLT (35 or 39F left-sided DLT) in the airway simulator (MedSim Eagle; MedSim Advanced Medical Simulation, Binghamton, New York, USA). The DLT was initially placed in the trachea under direct vision. The stylet was removed, and the tube was then rotated 90° towards the left and passed into what the operator believed was the proper position. The tracheal cuff was inflated. A fibreoptic bronchoscope was then introduced into the tracheal lumen to visualise the carina. The endobronchial cuff was then inflated with 2 cm3 of air, and the position of the cuff was visualised and adjusted until the edge of the blue cuff was located in the left mainstem bronchus, approximately 5 mm below the carina.5 The right mainstem bronchus was also identified. Because the tracheobronchial tree anatomy of the mannequin used in this study does not have the take off of the right upper bronchus or secondary bifurcation on the left mainstem bronchus, a picture was shown to each anaesthesiologist with a graphic display of the DLT in the optimal position along with a specific description of the origin of the right upper bronchus and left upper and left lower bronchus.

To increase the margin of safety while positioning a left-sided DLT, if the upper surface of the blue endobronchial cuff was not seen below the tracheal carina in the left mainstem bronchus, a radiopaque marker that is included in the body of the endobronchial lumen was suggested to the participant anaesthesiologist to serve as an alternative reference marker to ensure proper placement of the tube. This radiopaque marker appears white when reflected with the fibrescope and needs to be seen above the tracheal carina, similar to the landmarks recommended by Fortier et al.6

After the hands-on demonstration was concluded, the participating anaesthesiologists then practised tube placement until they achieved the optimal position of the DLT with the fibreoptic bronchoscope. Correct tube placement was confirmed by the supervisor (JHC or KU). The average time spent with the airway simulator was approximately 45 min per trainee.

Digital video disc group

The second group was provided with a DVD made by JHC, an experienced thoracic anaesthesiologist, which included a demonstration of the proper placement of a DLT, including the fibreoptic bronchoscopy view and a complete narration of the procedure. Anaesthesiologists assigned to this group reviewed the 14-min long DVD on their personal computer and played the DVD as many times as they wished (the minimal time reviewed among participants was one and maximum of two replays of the DVD; n = 30 patients received allocated intervention in this group).

Group assignments were made via computer-generated random number sequence. Numbered, opaque envelopes were prepared containing the group assignments for the participating anaesthesiologist. The random sequence and envelopes were created by Dr Hallam. Patients and faculty were enrolled by Dr Hallam. The envelope was opened by the participating anaesthesiologist the day before or several hours prior to anaesthetic induction, to provide enough time for training. Physicians recording data were blinded to which group the participating anaesthesiologist had been assigned. Patients were also blinded to their intervention group. Figure 1 displays the patients' enrolment and allocation within their respective group.

Fig. 1
Fig. 1


Two faculty anaesthesiologists (JHC and KU) were responsible for the care of all patients. These individuals managed the induction of the patients as well as all aspects of their care after placement of the DLT, but did not advise or supervise the participating anaesthesiologist during endotracheal tube placement. All patients received intravenous glycopyrrolate (0.2–0.4 mg) before surgery. After placement of standard monitors and radial arterial catheters, anaesthesia was induced with an intravenous agent and maintained with fentanyl and a volatile agent in oxygen. A non-depolarising muscle relaxant was used for paralysis.


Endotracheal intubation was performed with a left-sided Broncho-Cath (Mallinckrodt Medical; St Louis, Missouri, USA) in all patients. The selection of the proper DLT size was left to the participating anaesthesiologist. However, if input by JHC or KU was requested, an appropriate DLT tube size was suggested.

The styleted DLT was introduced into the glottis via direct laryngoscopy. Subsequent placement and positioning was done as described above in the simulator scenario. In each patient, a stopwatch was started as soon as the endotracheal tube passed the vocal cords. The stopwatch was stopped when the participating anaesthesiologist concluded that the tube was correctly placed; when the participating anaesthesiologist concluded that placement was not possible; or after a maximum of three placement attempts (each attempt was defined as withdrawing the bronchial lumen of the DLT into the trachea and then attempting to reposition it). After the participating anaesthesiologist had completed DLT placement, with patients still in the supine position, one of the supervising anaesthesiologists and primary investigator or co-investigator of the study with experience in fibreoptic bronchoscopy techniques and DLTs (JHC or KU) inserted a fibreoptic bronchoscope and determined whether proper positioning had been achieved. The criteria used to assess positioning failure were more than 50% herniation of the bronchial cuff herniated into carina; bronchial cuff edge not visible in the entrance of mainstem bronchus and radiopaque marker not seen above the tracheal carina, indicating that the cuff had likely been advanced to sufficient depth to occlude a secondary bronchus; DLT tube in the right mainstem bronchus; and unable to distinguish tracheal and bronchial anatomy. If the participating anaesthesiologist failed to successfully complete DLT placement, the supervising anaesthesiologist repositioned the tube properly. The amount of time to accomplish correct placement was also recorded.

We recorded the success of the DLT placement and the time required to complete the task. In addition, to indirectly assess the participating anaesthesiologist's skill with the fibreoptic bronchoscope, the number of times that the fibreoptic bronchoscope was passed through the endotracheal tube during positioning was recorded. To ensure that failures were not due to unique anatomical conditions, whenever a failure or malposition was identified, we recorded the time required by an experienced thoracic anaesthesiologist (JHC, KU) to correctly position the device.

When satisfactory device placement was achieved in the supine position, the endobronchial cuff of the DLT was deflated. After patients were in the lateral decubitus position, the endobronchial cuff was inflated, and the tube placement was reassessed via fibreoptic bronchoscopy.

Statistical analysis

On the basis of a previous study,1 the sample size was calculated to permit the detection of at least 2-min difference between any group in the time to device placement, with an α of 0.05 and power of 0.80.

Logistic regression analysis, which was fitted by the method of generalised estimating equations (GEEs), was used to test for differences in the probability of satisfactory placement between the intubation model simulator training group and the computer-based DVD group. As anaesthesiologists performed DLT placements on more than one patient (19 anaesthesiologists), the GEE method was used to account for the correlation of the tube placement outcomes by the same anaesthesiologist. In addition, the model also included the number of times an individual participated to adjust for possible effect of repetition. A similar analysis was done with the number of fibreoptic bronchoscopies performed (defined as number of separate insertions of the bronchoscope into the endotracheal tube).

For analysis of patient characteristics and demographics, a χ2 test was used and a P value less than 0.05 was considered statistically significant. Values are expressed as median, unless otherwise specified.


Sixty out of 72 patients were assigned to one of the instruction groups and received OLV (see Fig. 1). Patients were enrolled from March 2007 to December 2008. The trial was completed as planned. The surgical procedures performed included the following: 55 lung procedures and five oesophageal surgeries. The number of thoracoscopies and thoracotomies in each of the two training groups were not significantly different. The number of left-sided or right-sided procedures did not differ significantly. The patients intubated by non-thoracic anaesthesiologists in the two groups studied were statistically equivalent with regard to age and weight. The airway simulator patient group consisted of 14 men and 16 women. The computer-based DVD patient group consisted of 20 men and 10 women. The DLT sizes used in this study were 35 French (n = 10), 37 French (n = 18), 39 French (n = 20) and 41 French (n = 12).

There was no significant difference in the estimated probability of satisfactory double-lumen tube placement between the simulator training group [80.5%, 95% confidence interval (CI) 58.2–96.2%] and in the computer-based DVD training group (75.6%, 95% CI 49.8–88.5%; P = 0.378). The effect of number of times an individual participated was also not significant (P = 0.312). In addition, all participants (non-thoracic anaesthesiologist, n = 27) were exposed to either a simulator or computer-based training more than once; however, following the inclusion criteria of less than two lung isolation cases per month, only one individual in the simulator group and three in the computer-based DVD training group were withdrawn because their exposure to the training was less than a month apart. Therefore, the final statistics were based on 29 exposures in the simulator group and 27 in the computer-based DVD group. In fact, the statistical analysis with or without the withdrawn anaesthesiologists was similar in result. Figure 2 shows the success/failure rates for the two groups studied.

Fig. 2
Fig. 2

Participating anaesthesiologists from both groups failed to achieve proper positioning of the left-sided DLT in 14 of 60 patients studied (23%). In the intubation model simulator group, there were five of 30 (17%) failed placements, and in the computer-based DVD group, there were nine of 30 (30%) failed placements. There was no significant difference in the time required for positioning between the two groups (P = 0.406; Table 1). In failed studies, it took from 15 to 120 s for the thoracic anaesthesiologist (JHC or KU) to correct any malpositions. There was no significant difference in the number of total fibreoptic bronchoscopies between the two groups (P = 0.267; Table 2). Table 3 shows the reason for failed placement in each group. In order to ensure patient safety, the independent observer corrected all failed placements and lung collapse was successfully achieved in all participants studied. There were no recognised complications as a result of tube placement or OLV.

Table 1
Table 1:
Comparison of the time for positioning between intubation model simulator group and the computer-based digital video disc group
Table 2
Table 2:
Description of the number of fibreoptic bronchoscopies
Table 3
Table 3:
Malposition criteria and incidence of failures


Numerous studies have demonstrated the utility of simulators for assessment of performance and for crisis management7 or emergency airway management using scenario-based simulation training.8 Another potential role of simulators is for continuing education of postgraduate anaesthesiologists or trainees specialised in airway management.9 In this context, the airway simulator mannequin has potential advantages, including interactive discussion between the expert and the trainee; capability to practise an intubation technique and fibreoptic bronchoscopy; and ability to test and confirm optimal DLT placement.

In contrast, relative advantages of a DVD as a training tool include increased availability; repeatability; and easy access with most computers. Limited disadvantages include the need to make a professional DVD in addition to the expert's time; no interactive discussion; and no ‘hands-on’ practice with the bronchoscope.

In our current study, we found that the effectiveness of the intubation model simulator training versus the DVD training group was not statistically different. These two training methods were comparable in their net result – the successful placement of DLTs by non-thoracic anaesthesiologists. It should be noted that these findings may be attributed to the small number of study participants and not to the equality of the two training methods. Thus, the intubation model may be a preferable training method because it provides the hands-on experience that is necessary to learn the technical skills and allows the trainee to obtain optimal placement of DLTs with the aid of the fibrescope as many times as needed. Also, an experienced thoracic anaesthesiologist could facilitate training and advised on the correction of malpositions as needed.

The incidence of unrecognised malpositions overall was 23% in this study. This rate of DLT malposition is less than that observed in a historical control group (38% malpositions), wherein the participants received no prior training with the expert regarding placement of DLTs, but only a brief discussion. However, it is important to point out that the median time required to successfully place a left-sided DLT was 4.6 and 5.6 min, in the simulator and DVD groups, respectively. In a previous study, Campos and Kernstine10 reported that the average time required to place a left-sided DLT for an experienced thoracic anaesthesiologist was approximately 2 min. Therefore, although these training methods appear to generally improve performance, anaesthesiologists with limited thoracic experience still have a high level of failure and, even when successful, require considerably greater time to place DLTs. In this study, if the time required to place the DLT exceeded approximately 8–10 min, it was highly unlikely the DLT was going to be successfully placed.


One of the limitations of the study is that it was performed in a single institution and does not reflect the expertise of other thoracic or non-thoracic anaesthesiologists. In addition, the intubation model simulator mannequin does not have an ideal tracheobronchial tree, which might have influenced the knowledge of the tracheobronchial anatomy in some participants of the study. However, the anaesthesiologists who viewed the DVD, which showed in-vivo anatomy of the tracheobronchial tree, also had difficulty in recognising the anatomy when placing the tube. In one patient within the DVD group (1/30), despite the fact that the DLT was in a good position, the non-thoracic anaesthesiologist was unable to confirm proper placement. This is clearly demonstrated by the fact that in both groups, the participants had an increased number of bronchoscopies, from two to five attempts while placing DLTs. In addition, disadvantages of a mannequin include time spent by thoracic anaesthesiologist with the non-thoracic anaesthesiologist and the high cost of maintaining the device.

An expert in thoracic anaesthesia from our institution made the DVD; it represents the practice of one anaesthesiologist. To minimise this potential bias, guidelines by Benumof et al.5 for proper placement of left-sided DLT were followed. In addition, we believe that the radiopaque marker recommended by Fortier et al.6 made a difference while teaching the two methods used and this might explain in part the reduced incidence of unrecognised malpositions in the current study. To date, there is no study that defines experience necessary for proficiency in lung isolation techniques. Although the baseline level of experience for the non-thoracic anaesthesiologists studied regarding placement of DLTs was not assessed, many of them had not placed a DLT within months prior to the study. There is also no study to determine the effectiveness of teaching methods or which defines the relevant learning curve.

One recent study11 involving medical students, who were tested to perform fibreoptic intubation in a mannequin model, after an expert bronchoscopist provided a didactic lecture with, or without, virtual fibreoptic intubation (VFI) software, showed an increased success rate on hands-on practice when the VFI software was used as an adjunct compared to the expert bronchoscopist group alone. Perhaps, a similar teaching method could be recommended when training non-thoracic anaesthesiologists, so the teaching from the expert combined with the CD ROM material could improve the success rate of the placement of DLTs.

Our study showed an overall failure to place DLTs in 23% of the patients in the groups studied, which is a lower failure rate than we found previously.1 It appears therefore, that a refresher course prior to the case, whether via intubation model simulator mannequin or DVD instruction or a combination of both, can be used to enhance successful placement of DLTs by anaesthesiologists with limited thoracic anaesthesia experience. These methods of teaching should be considered in any institution that performs thoracic cases requiring lung isolation or in any emergency setting in which a non-specialist needs to place a DLT tube. Further studies are needed to assess the level of retention once the trainee has participated in training sessions.


The present work was supported by the Department of Anesthesia at The University of Iowa, Iowa City, Iowa, USA. None of the authors has any conflict of interest.


1 Campos JH, Hallam EA, Van Natta T, Kernstine KH. Devices for lung isolation used by anesthesiologists with limited experience: comparison of double-lumen endotracheal tubes, Univent torque control blocker and Arndt wire-guided endobronchial blocker. Anesthesiology 2006; 104:261–266.
2 Hesselfeldt R, Kristensen MS, Rasmussen LS. Evaluation of the airway of the SimMan full-scale patient simulator. Acta Anaesthesiol Scand 2005; 49:1339–1345.
3 Wong AK. Full scale computer simulators in anesthesia training and evaluation. Can J Anesth 2004; 51:455–464.
4 Nyssen AS, Larbuisson R, Janssens M, et al. A comparison of the training value of two types of anesthesia simulators: computer screen-based and mannequin-based simulators. Anesth Analg 2002; 94:1560–1565.
5 Benumof JL, Partridge BL, Salvatierra C, Keating J. Margin of safety in positioning modern double-lumen endotracheal tubes. Anesthesiology 1987; 67:729–738.
6 Fortier G, Coté D, Bergeron C, Bussieres JS. New landmarks improve the position of the left Broncho-Cath double-lumen tube: comparison with the classic technique. Can J Anesth 2001; 48:790–794.
7 Schwid HA, Rooke GA, Carline J, et al. Anesthesia Simulator Research Consortium. Evaluation of anesthesia residents using mannequin-based simulation. Anesthesiology 2002; 97:1434–1444.
8 Rosenthal ME, Adachi M, Ribaudo V, et al. Achieving housestaff competence in emergency airway management using scenario based simulation training: comparison of attending vs housestaff trainers. Chest 2006; 129:1453–1458.
9 Kory PD, Eisen LA, Adachi M, et al. Initial airway management skills of senior residents: simulation training compared with traditional training. Chest 2007; 132:1927–1931.
10 Campos JH, Kernstine KH. A comparison of left-sided Broncho-Cath with the torque control blocker Univent and the wire-guided blocker. Anesth Analg 2003; 96:283–289.
11 Boet S, Bould MD, Schaeffer R, et al. Learning fibreoptic intubation with a virtual computer program transfers to ‘hands on’ improvement. Eur J Anaesthesiol 2010; 27:31–35.

computer-based digital video disc; double-lumen endotracheal tube; fibreoptic bronchoscopy; intubation simulator; thoracic anaesthesia

© 2011 European Society of Anaesthesiology