Learning Curve Analysis of Double-Lumen Tube Placement in Anesthesiology Residents: The Impact of Prior Experience and Laryngoscope Type

Learning Curve Analysis of Double-Lumen Tube Placement in Anesthesiology Residents: The Impact of Prior Experience and Laryngoscope Type

Placement of a double-lumen tube (DLT) is one of the fundamental techniques for safely achieving one-lung ventilation in thoracic anesthesia. Compared with standard single-lumen endotracheal tubes, DLTs have a larger diameter, greater length, and a more rigid structure. These characteristics make intubation with a DLT considerably more challenging in both normal and difficult airways, potentially increasing the risk of upper airway trauma and postoperative complications, thereby contributing to increased overall morbidity and mortality.

Despite its widespread clinical use, DLT placement remains a technically demanding procedure due to anatomical variability, limited visualization, and operator-dependent performance differences. The reported incidence of DLT malposition in the literature ranges from 32% to 83%. Even in studies reporting lower rates (approximately 32.44%), it is acknowledged that a substantial number of malpositions may go undetected, which can critically impact both patient safety and surgical outcomes. In addition to initial misplacement, repeated repositioning of a malpositioned DLT may itself lead to airway injury. Increased manipulation of the DLT within the bronchial tree may result in excessive trauma.

DLT malposition may predispose patients to inadequate lung isolation, hypoxemia, tracheobronchial injury, and potential complications such as barotrauma or pneumothorax. Operator experience is recognized as a key determinant of success in airway management. Previous studies have demonstrated that technical proficiency in advanced airway interventions improves with cumulative clinical experience. However, most existing research has focused on basic airway devices, and evidence regarding advanced airway techniques such as DLT placement remains limited. Furthermore, the relative contributions of operator experience, airway difficulty, and device technology to procedural success have not been fully elucidated.

To address the need for easier and safer endotracheal intubation, various types of video laryngoscopes have been developed as alternatives to direct laryngoscopy. Nevertheless, it remains unclear whether video-assisted airway management reduces DLT malposition rates across different levels of operator experience.

Simulation-based training has become increasingly important in modern anesthesiology education, providing a safe and standardized environment for the assessment of airway management skills. Accordingly, there is a need for simulation-based studies evaluating learning curves and technical performance in DLT placement.

The primary aim of this study is to evaluate the effects of overall clinical experience and prior DLT placement experience on malposition rates during DLT insertion procedures performed in a simulation setting among anesthesiology residents. Secondary objectives include comparing the performance of video laryngoscopy and Macintosh laryngoscopy, as well as analyzing learning curves across repeated attempts. Additionally, the interaction between experience level, type of laryngoscope used, and airway difficulty will be investigated.

Study Design and Population This study will be conducted in accordance with the Declaration of Helsinki and will be carried out at Ankara Atatürk Sanatorium Training and Research Hospital after obtaining approval from the institutional ethics committee.

The study is designed as a prospective, observer-blinded, simulation-based study with repeated measures. It aims to evaluate DLT placement performance using standardized airway simulation models.

All anesthesiology residents working in the Department of Anesthesiology and Reanimation at Ankara Atatürk Sanatorium Training and Research Hospital will be invited to participate on a voluntary basis (total of 33 participants). Written informed consent will be obtained from all participants.

Anesthesiology residents with varying levels of experience will be included voluntarily. Participants will be stratified based on years of residency training and the number of prior DLT placements performed.

Airway simulation will be conducted using high-fidelity airway manikins equipped with bronchial anatomy modules. Intubation will be performed using a 37 Fr DLT. Simulation platforms with fiberoptic verification capability will be utilized. Prior to data collection, participants will receive standardized training on DLT placement.

Training Provided to All Participants:

• One standardized theoretical training session
• One video demonstration
• One practice attempt for each scenario (not included in the data analysis)

The following variables will be recorded:

• Age of the participant
• Sex
• Year of residency training
• Number of prior real-life DLT intubations performed
• Intubation time
• Number of attempts
• Presence of malposition
• Visual Analog Scale (VAS) difficulty score (participants will rate the difficulty of each attempt on a scale from 0 = very easy to 100 = very difficult)

Each participant will perform the following four clinical scenarios in a randomized order:

1. Scenario: Normal airway - Macintosh laryngoscope
2. Scenario: Normal airway - Video laryngoscope
3. Scenario: Difficult airway - Macintosh laryngoscope
4. Scenario: Difficult airway - Video laryngoscope Participants will complete the scenarios in different sequences, determined by computer-based randomization.

For each participant:

• Each scenario will be performed three times (three repetitions).
• Adequate rest periods will be provided between repetitions to minimize fatigue (at least 10 minutes between repetitions).
• A maximum of two intubation attempts will be allowed per repetition.
• Participants will perform no more than two scenarios per day.
• DLT placement will be confirmed using fiberoptic bronchoscopy by two independent anesthesiologists who are blinded to the identity and experience level of the operator. Interobserver agreement will be assessed using Cohen's kappa analysis.

If successful intubation is not achieved after two attempts, the procedure will be recorded as a failure for that repetition. The presence of malposition will be recorded after each attempt. The total number of malpositions observed across all procedures and repetitions will be documented.

Definition of Malposition:

Malposition is defined as any of the following:

• Failure of the bronchial cuff to be positioned in the left main bronchus, remaining instead in the trachea
• Unintentional placement of the DLT into the right bronchial system
• Advancement of the DLT at least 1 cm beyond the optimal position within the left main bronchus Statistical Analysis Statistical analyses will be performed using SPSS Version 27 (IBM Corp., Armonk, NY, USA), R software (R Foundation for Statistical Computing, Vienna, Austria), and Microsoft Excel (Microsoft Corp., USA). Descriptive statistics will be presented as mean ± standard deviation or median (interquartile range), as appropriate. Categorical variables will be expressed as counts and percentages. The normality of data distribution will be assessed using the Shapiro-Wilk test.

Given the repeated-measures structure of the data, mixed-effects models will be employed for analysis. Learning curve analyses will be conducted using the Cumulative Sum Control Chart (CUSUM) method and regression-based slope analyses. A p-value of < 0.05 will be considered statistically significant.

Sample Size In this study, the sample size was determined using an event-based approach, taking into account the lack of directly comparable studies in the literature and the limitations of conventional sample size calculation methods for multilevel data with repeated measures. A total of 33 participants, 4 scenarios, and 3 repetitions are planned, yielding 396 measurements.

Based on the lowest reported malposition rate in the literature (32%), approximately 126 events (malpositions) are expected. According to the commonly recommended rule of at least 10 events per variable in multivariable regression analyses, a minimum of 50 events would be required for approximately 5 variables to be included in the model. Therefore, the anticipated number of events is considered more than sufficient, and the planned sample size is deemed adequate to test the study hypotheses.

Inclusion Criteria:

• Being an anesthesiology resident at the study institution
• Voluntary participation

Exclusion Criteria:

• Refusal to participate in the study
• Prior participation in the same simulation scenario