Fiberoptic Orotracheal Intubation on Anesthetized Patients: Do Manipulation Skills Learned on a Simple Model Transfer into the Operating Room?

Background: With increasing pressure to use operating room time efficiently, opportunities for residents to learn fiberoptic orotracheal intubation in the operating room have declined. The purpose of this study was to determine whether fiberoptic orotracheal intubation skills learned outside the operating room on a simple model could be transferred into the clinical setting.

Methods: First-year anesthesiology residents and first- and second-year internal medicine residents were recruited. Subjects were randomized to a didactic-teaching-only group (n = 12) or a model-training group (n = 12). The didactic-teaching group received a detailed lecture from an expert bronchoscopist. The model-training group was guided, by experts, through tasks performed on a simple model designed to refine fiberoptic manipulation skills. After the training session, subjects performed a fiberoptic orotracheal intubation on healthy, consenting, anesthetized, paralyzed female patients undergoing elective surgery with predicted “easy” laryngoscopic intubations. Two blinded anesthesiologists evaluated each subject.

Results: After the training session, the model group significantly outperformed the didactic group in the operating room when evaluated with a global rating scale (P < 0.01) and checklist (P < 0.05). Model-trained subjects completed the fiberoptic orotracheal intubation significantly faster than didactic-trained subjects (P < 0.01). Model-trained subjects were also more successful at achieving tracheal intubation than the didactic group (P < 0.005).

Conclusion: Fiberoptic orotracheal intubation skills training on a simple model is more effective than conventional didactic instruction for transfer to the clinical setting. Incorporating an extraoperative model into the training of fiberoptic orotracheal intubation may greatly reduce the time and pressures that accompany teaching this skill in the operating room.

SECURING the airway of a patient with a known or anticipated difficult tracheal intubation is usually performed with a flexible fiberoptic bronchoscope while the patient is conscious and breathing spontaneously. ¹

Fiberoptic orotracheal intubation (FOI) is an advanced airway skill and, as such, requires extra training. However, the number of patients with anticipated difficult laryngoscopic tracheal intubation is limited; therefore, anesthesiology residents may not receive adequate FOI experience during their training. ²

There is growing acceptance that teaching FOI on anesthetized and paralyzed patients with predicted or documented easy laryngoscopic tracheal intubation will lead to improved proficiency when dealing with abnormal airways in the awake state. ³ The acquisition of this skill on anesthetized patients with normal airway anatomy should also be beneficial when abnormal airways are encountered unexpectedly after induction of anesthesia and administration of a neuromuscular relaxant. ⁴

Traditionally, FOI training on anesthetized patients with normal airway anatomy has occurred in the operating room (OR) under the guidance of an attending anesthesiologist experienced in the skill. Fiscal restraints have resulted in pressure to keep “turnover” in the OR high. Consequently, there is less time for attending staff to teach this technical skill and for residents to practice this skill under expert supervision. Furthermore, learning FOI on patients usually occurs without previous “hands-on” practice. Without previous experience in handling the fiberoptic bronchoscope, the potential for complications and patient harm may be increased. Thus, debate continues as to whether the acquisition of technical skills on actual patients is an ethical practice. ⁵ For these reasons, educators have been forced to examine other methods to teach this technical skill.

“Choose-the-hole” models enhance manipulation skills with the fiberoptic bronchoscope by reinforcing proper directional handling of the fiberoptic bronchoscope as it is guided under fiberoptic vision through a maze. ^4,6 These models can be as simple as a series of wooden panels with holes or as detailed as anatomically correct airway mannequins. ^4,6,7 If manipulation skills are important to a successful FOI, then practice on these models should improve FOI performance. To date, no studies have objectively demonstrated the transfer of skills learned on bench models to humans in the OR with valid and reliable evaluation instruments; specifically, global assessment scales and checklists. ⁸

The purpose of this study was to assess whether fiberoptic bronchoscopic manipulation skills learned on a simple extraoperative model could be transferred into the clinical setting and improve the trainee’s ability to perform an FOI.

Material and Methods

After obtaining approval from the University of Toronto ethics committee (Toronto, Ontario, Canada), first-year anesthesiology residents and first- and second-year internal medicine residents were recruited as subjects, and informed consent was obtained. All subjects were asked to complete a questionnaire to assess their previous exposure to tracheal intubation and flexible endoscopy. Subjects who had performed more than five independent nasopharyngoscopies, lower gastrointestinal endoscopies, flexible arthroscopies, or bronchoscopies via an endotracheal tube were excluded. Subjects who had performed more than three FOIs independently were also excluded.

Subjects received a manual that described in detail the airway and fiberoptic bronchoscope anatomy. The manual also contained figures and step-by-step instruction on how to perform an FOI on an anesthetized and paralyzed patient. Subjects were asked to review this manual before attending the training session. In addition, all subjects viewed a 20-min video that further emphasized the anatomy of the fiberoptic bronchoscope and handling of the instrument. The video concluded by demonstrating an FOI on an anesthetized and paralyzed patient and provided simultaneous ex vivo and in vivo camera angles with verbal commentary.

Subjects were randomized into one of two groups: model training or didactic training. All subjects proceeded to familiarize themselves with the preparation and handling of a fiberoptic bronchoscope during a 10-min instrument orientation by three expert bronchoscopists. All testing and training was performed with a 5-mm Pentax fiberoptic bronchoscope (model no. FB-15BS; Misssissauga, Ontario, Canada) and video monitor set-up.

Fig. 1

Subjects were pretested using a previously described “choose-the-hole” model designed to refine fiberoptic bronchoscope manipulation skills (fig. 1). ⁴ The pretest task involved manipulating the fiberoptic bronchoscope through a specified syringe barrel combination in the covered wooden model under fiberoptic vision. A blinded anesthesiologist evaluated each subject on a text-anchored global assessment scale (scored from 1 to 5) on their ability to handle the instrument (Appendix 1). Time to complete the task was also measured.

Subjects were trained according to their randomization. The didactic-training group received a 45-min lecture by an expert bronchoscopist that emphasized proper handling and usage of the fiberoptic bronchoscope for FOI. The model-training group spent 45 min refining their fiberoptic bronchoscope manipulation skills, under the guidance of experts, on the choose-the-hole model by practicing different syringe barrel combinations (fig. 1). Both groups were taught how to load a polyvinylchloride endotracheal tube over the fiberoptic bronchoscope and to advance the endotracheal tube into the trachea with the bevel facing posteriorly once the carina is visualized. ⁹ Immediately after their respective training sessions, both groups were posttested on the pretest model to ensure improvement before OR assessment. If subjects did not significantly improve from their pretest scores and times, the study would be terminated. In the 10 days after the training session, subjects were tested in the OR on their ability to perform an FOI on an anesthetized and paralyzed patient.

After obtaining University of Toronto ethics approval (Toronto, Ontario, Canada), consent was obtained from patients in the preadmission unit before their proposed elective surgery. Female patients requiring tracheal intubation for their elective surgery were recruited. Only female patients were approached in an attempt to reduce variability during tests in the OR. Exclusion criteria included allergies to medications used, risk for pulmonary aspiration, American Society of Anesthesiologists classification greater than II, and a potential difficult airway to allow for a safe alternative method of tracheal intubation if FOI failed. ^10,11

All patients were monitored continuously with pulse oximetry, a five-lead electrocardiogram, and a noninvasive blood pressure cuff. Three anesthesiologists were present for all study cases: two for blinded assessment of each subject, and one to manage the patient’s anesthetic. The study investigator intervened if blood pressure or heart rate was not maintained within 20% of the patient’s baseline readings, pulse oximetry decreased below 94%, or the maximum 210 s had expired. Through pilot work, we determined that 210 s provided a safe margin for respiratory and hemodynamic stability in healthy apneic female patients. Experts were able to achieve tracheal intubation with the fiberoptic bronchoscope in this patient population within 30 s on average.

Patients received 0.2 mg glycopyrrolate intravenously as an antisialogue on establishing intravenous access in the OR. Patients were then preoxygenated with 100% oxygen for 3 min. General anesthesia with muscle relaxation was induced with 1 mg midazolam, 2 μg/kg fentanyl, 2.5 mg/kg propofol, and 0.6 mg/kg rocuronium. Patients were manually ventilated with 100% oxygen to an end-tidal carbon dioxide between 25 and 35 mmHg until loss of twitch response by nerve stimulation. A trained OR respiratory therapist assisted each trial by providing neck extension, jaw thrust, and lingual traction. ^12–14 Each subject then attempted tracheal intubation using the fiberoptic bronchoscope and video monitor set-up. Additional propofol (0.75 mg/kg) was administered as necessary to maintain general anesthesia. If tracheal intubation was unsuccessful within 210 seconds or if there was oxygen desaturation less than 94%, the fiberoptic bronchoscope was removed, and the subject was automatically considered a treatment failure. After 1 min of manual ventilation with 100% oxygen, the study investigator performed the FOI.

A successful tracheal intubation was defined as fiberoptic bronchoscopic confirmation of the carina after placement of the endotracheal tube in the allotted 210 s. Two blinded consultant anesthesiologists completed a previously validated checklist scored from 1 to 10 during each subject’s FOI attempt (Appendix 2). Examiners also evaluated FOI performance using a validated global rating assessment scale scored from 8 to 40 (Appendix 3). ¹⁵ Examiners were also asked to give each subject an overall pass or fail rating at the end of each trial. A pass rating was given if evaluators felt the subject could perform a second FOI with no additional training. Thus, a subject could still receive a fail rating if an examiner thought they required further training regardless of whether they achieved tracheal intubation. One of the two evaluators was common for all subjects. One of three anesthesiologists was chosen as the second evaluator for each subject.

Subjects were timed during the OR test. Timing commenced when the fiberoptic bronchoscope first entered the oropharynx and stopped when the fiberoptic bronchoscope was removed from the patient after passing the endotracheal tube and confirming its correct placement by visualization of the carina.

Statistical Analysis

Statistical analysis was performed using SPSS 10.0 (Chicago, IL). In a pilot study, we examined the effects of model training on ureteroscopic skills and found a significant effect of training using genitourinary bench models when compared with a control group receiving didactic instruction only. The effect size of training using the bench model compared with the control group was 1.92 SDs using a global rating scale. Borrowing from the psychological field, effect sizes greater than 1.0 SDs are acceptable in assessing teaching interventions. ¹⁶ With 12 subjects in each group, using a β of 0.20 and a two-tailed α of 0.05, we had 80% power to detect an effect size of 1.2 SDs.

Categoric data including success–failure and pass–fail results were analyzed by the chi-square test. Time to complete the tasks (in seconds) and time between training and OR assessment (in days) were analyzed using Mann–Whitney U tests. Parametric data, including global rating scores and checklist scores, were compared using independent groups t tests. Interrater reliability for the global rating and checklist scores were assessed using the Pearson correlation coefficient. A two-tailed P value less than 0.05 was considered statistically significant.

Results

Table 1

Subject demographics, including previous experience, pretest performance, and number of days between training and the OR test, were not different between the two groups (table 1). Twelve subjects were enrolled in the didactic-training group, and 12 were enrolled in the model-training group. After their respective training sessions, both groups improved significantly on the posttest from their pretest global assessments (P < 0.001) and pretest times (P < 0.05). The model group tended toward greater improvement from their pretest scores and times than the didactic group (P = 0.06).

Fig. 2

Table 2

Subjects in the model group received higher global rating assessment scores (P < 0.01) and higher checklist scores (P < 0.05) than subjects in the didactic group (fig. 2). Evaluators passed more subjects who received model training (75%) than didactic training (33%;P < 0.05) on their FOI performance (table 2).

Subjects who received model training achieved a successful tracheal intubation (92%) using the fiberoptic bronchoscope more often than those who received didactic training (42%;P < 0.005;table 2). Only one subject in the model group was unable to perform a successful tracheal intubation. All failures were the result of expired time. No patients suffered desaturations less than 94% or hemodynamic instability. All failures were subsequently successfully intubated easily with a fiberoptic bronchoscope by a study investigator.

Model-trained subjects performed the task faster (median, 81 s) than didactic-trained subjects (median = 210 s;P < 0.01;table 2).

Intraoperative interrater reliability was strong for global rating assessments (r = 0.78–0.84;P < 0.05) and very strong for checklist scores (r = 0.87–0.99;P < 0.01).

Discussion

Educators have sought alternative methods of introducing and teaching technical skills to residents to maximize the experience in the clinical setting. Extraoperative courses have been used by different specialties to improve exposure to various technical skills. Most training courses rely heavily on bench models to simulate the intraoperative experience.

Surgical educators have examined the benefits of technical skills training in the laboratory setting on bench models with instruments such as the Objective Structured Assessment of Technical Skills. ^15,17,18 They found that assessment of a technical skill using a global rating scale and checklist was a highly valid and reliable method of evaluating proficiency. Moreover, Martin et al.¹⁷ used the Objective Structured Assessment of Technical Skills to compare testing skills on bench models and live animals as human surrogates. They found that model simulation yielded equivalent results to the use of live animals. Anastakis et al.⁸ furthered this work by demonstrating that skills learned on bench models could be transferred to human cadavers using similar validated evaluation methods. This study suggested that surgical skills learned on bench models could be transferred to humans in the OR. Our investigation confirms that a technical skill learned on a bench model can be transferred to patients in the clinical setting when evaluated with validated global rating assessments and checklists.

The results of this study confirmed that both hands-on model training and didactic training are beneficial. Both groups improved on their fiberoptic bronchoscope manipulation ability from their pretest evaluation to the posttest. Although the model group tended toward greater improvement, the result was not statistically significant. The carryover knowledge of the task in the control group from the pretest combined with the short interval between pretest and posttest may have contributed to the lack of statistical significance. ¹⁹ Furthermore, progressive error from physical fatigue after training in the model group may have impacted their posttest scores. ¹⁹

However, the intraoperative assessment indicated that training FOI on the relatively simple model did transfer with significantly greater efficacy into the OR than conventional didactic instruction. Subjects who were trained on the model received significantly better global assessment and checklist scores than their didactic counterparts. Furthermore, evaluators passed more subjects on the OR test who received model training than didactic training. Lastly, the model group completed the task more successfully and faster than the didactic group.

Despite our results, using simple choose-the-hole models as sole training of FOI on anesthetized and paralyzed patients is not appropriate. A successful FOI training program must also include intraoperative experience. Proficiency of this necessary skill is gained by experience and practice on appropriate patients with concurrent constructive feedback. ^3,20 In a recent review of airway management, Crosby et al.²¹ suggested that experience with a specific airway device is just as important as the device itself. ²¹ Models can serve as a useful and proven adjunct to FOI training by introducing and refining the fiberoptic bronchoscope manipulation skills that are required for proficiency in the clinical setting. Incorporating extraoperative models into the training of FOI may greatly reduce the time and pressures that accompany teaching this skill for the first time in the OR.

The authors thank the Surgical Skills Centre, Mt. Sinai Hospital, University of Toronto (Toronto, Ontario, Canada), and Pentax Canada (Misssissauga, Ontario, Canada) for resources during training and testing; Glenn Regehr, Ph.D. (Associate Professor, Centre for Research in Education, University of Toronto, Toronto, Ontario, Canada), for his valuable contributions; and Arthur Frederick David Cole, M.D. (Assistant Professor, University of Toronto, Toronto, Ontario, Canada), for the use of the model he designed for fiberoptic intubation training.