Standard clinical risk factors for difficult laryngoscopy are not independent predictors of intubation success with the GlideScope

Deep learning-based facial analysis for predicting difficult videolaryngoscopy: a feasibility study

While videolaryngoscopy has resulted in better overall success rates of tracheal intubation, airway assessment is still an important prerequisite for safe airway management. This study aimed to create an artificial intelligence model to identify difficult videolaryngoscopy using a neural network. Baseline characteristics, medical history, bedside examination and seven facial images were included as predictor variables. ResNet-18 was introduced to recognise images and extract features. Different machine learning algorithms were utilised to develop predictive models. A videolaryngoscopy view of Cormack-Lehane grade of 1 or 2 was classified as 'non-difficult', while grade 3 or 4 was classified as 'difficult'. A total of 5849 patients were included, of whom 5335 had non-difficult and 514 had difficult videolaryngoscopy. The facial model (only including facial images) using the Light Gradient Boosting Machine algorithm showed the highest area under the curve (95%CI) of 0.779 (0.733-0.825) with a sensitivity (95%CI) of 0.757 (0.650-0.845) and specificity (95%CI) of 0.721 (0.626-0.794) in the test set. Compared with bedside examination and multivariate scores (El-Ganzouri and Wilson), the facial model had significantly higher predictive performance (p < 0.001). Artificial intelligence-based facial analysis is a feasible technique for predicting difficulty during videolaryngoscopy, and the model developed using neural networks has higher predictive performance than traditional methods.

Development and validation of a regression model with nomogram for difficult video laryngoscopy in Chinese population: a prospective, single-center, and nested case-control study

Background

Airway management failure is associated with increased perioperative morbidity and mortality. Airway-related complications can be significantly reduced if difficult laryngoscopy is predicted with high accuracy. Currently, there are no large-sample studies on difficult airway assessments in Chinese populations. An airway assessment model based on the Chinese population is urgently needed to guide airway rescue strategy.

Methods

This prospective nested case-control study took place in a tertiary hospital in Shanghai, China. Information on 10,549 patients was collected, and 8,375 patients were enrolled, including 7,676 patients who underwent successful laryngoscopy and 699 patients who underwent difficult laryngoscopy. The baseline characteristics, medical history, and bedside examinations were included as predictor variables. Laryngoscopy was defined as 'successful laryngoscopy' based on a Cormack-Lehane Grades of 1-2 and as 'difficult laryngoscopy' based on a Cormack-Lehane Grades of 3-4. A model was developed by incorporating risk factors and was presented in the form of a nomogram by univariate logistic regression, least absolute shrinkage and selection operator, and stepwise logistic regression. The main outcome measures were area under the curve (AUC), sensitivity, and specificity of the predictive model.

Result

The AUC value of the prediction model was 0.807 (95% confidence interval [CI]: 0.787-0.828), with a sensitivity of 0.730 (95% CI, 0.690-0.769) and a specificity of 0.730 (95% CI, 0.718-0.742) in the training set. The AUC value of the prediction model was 0.829 (95% CI, 0.800-0.857), with a sensitivity of 0.784 (95% CI, 0.73-0.838) and a specificity of 0.722 (95% CI, 0.704-0.740) in the validation set.

Conclusion

Our model had accurate predictive performance, good clinical utility, and good robustness for difficult laryngoscopy in the Chinese population.

Glidescope Video Laryngoscopy in Patients with Severely Restricted Mouth Opening-A Pilot Study

BACKGROUND

An inter-incisor gap <3 cm is considered critical for videolaryngoscopy. It is unknown if new generation GlideScope Spectrum™ videolaryngoscopes with low-profile hyperangulated blades might facilitate safe tracheal intubation in these patients. This prospective pilot study aims to evaluate feasibility and safety of GlideScopeTM videolaryngoscopes in severely restricted mouth opening.

METHODS

Feasibility study in 30 adults with inter-incisor gaps between 1.0 and 3.0 cm scheduled for ENT or maxillofacial surgery. Individuals at risk for aspiration or rapid desaturation were excluded.

RESULTS

The mean mouth opening was 2.2 ± 0.5 cm (range 1.1-3.0 cm). First attempt success rate was 90% and overall success was 100%. A glottis view grade 1 or 2a was achieved in all patients. Nasotracheal intubation was particularly difficult if Magill forceps were required (n = 4). Intubation time differed between orotracheal (n = 9; 33 (25; 39) s) and nasotracheal (n = 21; 55 (38; 94) s); p = 0.049 intubations. The airway operator's subjective ratings on visual analogue scales (0-100) revealed that tube placement was more difficult in individuals with an inter-incisor gap <2.0 cm (n = 10; 35 (29; 54)) versus ≥2.0 cm (n = 20; 20 (10; 30)), p = 0.007, while quality of glottis exposure did not differ.

CONCLUSIONS

GlidescopeTM videolaryngoscopy is feasible and safe in patients with severely restricted mouth opening if given limitations are respected.

Creation of an artificial intelligence model for intubation difficulty classification by deep learning (convolutional neural network) using face images: an observational study

Background

Tracheal intubation is the gold standard for securing the airway, and it is not uncommon to encounter intubation difficulties in intensive care units and emergency rooms. Currently, there is a need for an objective measure to assess intubation difficulties in emergency situations by physicians, residents, and paramedics who are unfamiliar with tracheal intubation. Artificial intelligence (AI) is currently used in medical imaging owing to advanced performance. We aimed to create an AI model to classify intubation difficulties from the patient’s facial image using a convolutional neural network (CNN), which links the facial image with the actual difficulty of intubation.

Methods

Patients scheduled for surgery at Yamagata University Hospital between April and August 2020 were enrolled. Patients who underwent surgery with altered facial appearance, surgery with altered range of motion in the neck, or intubation performed by a physician with less than 3 years of anesthesia experience were excluded. Sixteen different facial images were obtained from the patients since the day after surgery. All images were judged as “Easy”/“Difficult” by an anesthesiologist, and an AI classification model was created using deep learning by linking the patient’s facial image and the intubation difficulty. Receiver operating characteristic curves of actual intubation difficulty and AI model were developed, and sensitivity, specificity, and area under the curve (AUC) were calculated; median AUC was used as the result. Class activation heat maps were used to visualize how the AI model classifies intubation difficulties.

Results

The best AI model for classifying intubation difficulties from 16 different images was generated in the supine-side-closed mouth-base position. The accuracy was 80.5%; sensitivity, 81.8%; specificity, 83.3%; AUC, 0.864; and 95% confidence interval, [0.731-0.969], indicating that the class activation heat map was concentrated around the neck regardless of the background; the AI model recognized facial contours and identified intubation difficulties.

Conclusion

This is the first study to apply deep learning (CNN) to classify intubation difficulties using an AI model. We could create an AI model with an AUC of 0.864. Our AI model may be useful for tracheal intubation performed by inexperienced medical staff in emergency situations or under general anesthesia.

Canadian Airway Focus Group updated consensus-based recommendations for management of the difficult airway: part 2. Planning and implementing safe management of the patient with an anticipated difficult airway

PURPOSE

Since the last Canadian Airway Focus Group (CAFG) guidelines were published in 2013, the published airway management literature has expanded substantially. The CAFG therefore re-convened to examine this literature and update practice recommendations. This second of two articles addresses airway evaluation, decision-making, and safe implementation of an airway management strategy when difficulty is anticipated.

SOURCE

Canadian Airway Focus Group members, including anesthesia, emergency medicine, and critical care physicians were assigned topics to search. Searches were run in the Medline, EMBASE, Cochrane Central Register of Controlled Trials, and CINAHL databases. Results were presented to the group and discussed during video conferences every two weeks from April 2018 to July 2020. These CAFG recommendations are based on the best available published evidence. Where high-quality evidence is lacking, statements are based on group consensus.

FINDINGS AND KEY RECOMMENDATIONS

Prior to airway management, a documented strategy should be formulated for every patient, based on airway evaluation. Bedside examination should seek predictors of difficulty with face-mask ventilation (FMV), tracheal intubation using video- or direct laryngoscopy (VL or DL), supraglottic airway use, as well as emergency front of neck airway access. Patient physiology and contextual issues should also be assessed. Predicted difficulty should prompt careful decision-making on how most safely to proceed with airway management. Awake tracheal intubation may provide an extra margin of safety when impossible VL or DL is predicted, when difficulty is predicted with more than one mode of airway management (e.g., tracheal intubation and FMV), or when predicted difficulty coincides with significant physiologic or contextual issues. If managing the patient after the induction of general anesthesia despite predicted difficulty, team briefing should include triggers for moving from one technique to the next, expert assistance should be sourced, and required equipment should be present. Unanticipated difficulty with airway management can always occur, so the airway manager should have a strategy for difficulty occurring in every patient, and the institution must make difficult airway equipment readily available. Tracheal extubation of the at-risk patient must also be carefully planned, including assessment of the patient's tolerance for withdrawal of airway support and whether re-intubation might be difficult.

Difficult Airway Characteristics Associated with First-Attempt Failure at Intubation Using Video Laryngoscopy in the Intensive Care Unit

RATIONALE

Video laryngoscopy has overcome the need to align the anatomic axes to obtain a view of the glottic opening to place a tracheal tube. However, despite this advantage, a large number of attempts are unsuccessful. There are no existing data on anatomic characteristics in critically ill patients associated with a failed first attempt at laryngoscopy when using video laryngoscopy.

OBJECTIVES

To identify characteristics associated with first-attempt failure at intubation when using video laryngoscopy in the intensive care unit (ICU).

METHODS

This is an observational study of 906 consecutive patients intubated in the ICU with a video laryngoscope between January 2012 and January 2016 in a single-center academic medical ICU. After each intubation, the operator completed a data collection form, which included information on difficult airway characteristics, device used, and outcome of each attempt. Multivariable regression models were constructed to determine the difficult airway characteristics associated with a failed first attempt at intubation.

MEASUREMENTS AND MAIN RESULTS

There were no significant differences in sex, age, reason for intubation, or device used between first-attempt failures and first-attempt successes. First-attempt successes more commonly reported no difficult airway characteristics were present (23.9%; 95% confidence interval [CI], 20.7-27.0% vs. 13.3%; 95% CI, 8.0-18.8%). In logistic regression analysis of the entire 906-patient database, blood in the airway (odds ratio [OR], 2.63; 95% CI, 1.64-4.20), airway edema (OR, 2.85; 95% CI, 1.48-5.45), and obesity (OR, 1.59; 95% CI, 1.08-2.32) were significantly associated with first-attempt failure. Data collection on limited mouth opening and secretions began after the first 133 intubations, and we fit a second logistic model to examine cases in which these additional difficult airway characteristics were collected. In this subset (n = 773), the presence of blood (OR, 2.73; 95% CI, 1.60-4.64), cervical immobility (OR, 3.34; 95% CI, 1.28-8.72), and airway edema (OR, 3.10; 95% CI, 1.42-6.70) were associated with first-attempt failure.

CONCLUSIONS

In this single-center study, presence of blood in the airway, airway edema, cervical immobility, and obesity are associated with higher odds of first-attempt failure, when intubation was performed with video laryngoscopy in an ICU.

Predictors of difficult intubation and the otolaryngology perioperative consult

Airway management is one of the most important aspects of anesthesia care. Although the incidence of difficult intubation is low, predicting a potentially difficult airway can ensure that necessary staff and equipment are available. A preoperative airway evaluation should include a history and physical examination focusing on elements that can cause problems with intubation. When indicated, flexible fiberoptic laryngoscopy can add valuable information regarding the upper aerodigestive anatomy. Specific patient and situational factors should be considered. Alternative plans should be defined before the initiation of anesthesia. Management of a complex airway should be a coordinated effort between anesthesiologists and otolaryngologists.

Validation of modified Mallampati test with addition of thyromental distance and sternomental distance to predict difficult endotracheal intubation in adults

Background and Aims:

Intubation is often a challenge for anaesthesiologists. Many parameters assist to predict difficult intubation. The present study was undertaken to assess the validity of different parameters in predicting difficult intubation for general anaesthesia (GA) in adults and effect of combining the parameters on the validity.

Methods:

The anaesthesiologist assessed oropharynx of 135 adult patients. Modified Mallampati test (MMT) was used and the thyromental distance (TMD) and sternomental distances (SMD) for each of the patients were also measured. The Cormack and Lehane laryngoscopic grading was assessed following laryngoscopy. The validity parameters such as sensitivity, specificity, false positive and negatives values, positive and negative predictive values were calculated. The effect of combining different measurements on the validity was also studied. Univariate analysis was performed using the parametric method.

Results:

The study group comprised of 135 patients. The sensitivity and specificity of MMT were 28.6% and 93%, respectively. The TMD (<6.5 CM) had sensitivity and specificity of 100% and 75.8%, respectively. The SMD (<12.5 CM) had sensitivity and specificity of 91% and 92.7%, respectively. Combination of MMT grading and TMD and SMD measurements increased the validity (sensitivity of 100% and specificity of 92.7%).

Conclusion:

MMT had high specificity. The validity of combination of MMT, SMD and TMD as compared to MMT alone was very high in predicting difficult intubation in adult patients. All parameters should be used in assessing an adult patient for surgery under GA.