This study shows that the DST, DTE and DSE can be accurately measured by ultrasound in the parasagittal plane. While we found significant differences for each of these measurements between the easy and difficult laryngoscopy groups (Table 1), we found that the DSE was a particularly successful independent predictor of difficult laryngoscopies as evaluated by logistic regression. Furthermore, we found that various other physiological measurements played a role in optimizing the predictive power of difficult laryngoscopies. In addition to the DSE, the best predictive model included such parameters as sex, BMI, and MMT. While the utility of the MMT was expected, as it is a direct visual measurement of airway opening, we found it interesting that the other factors of sex and BMI also contributed to this optimal model, as we discuss in subsequent paragraphs. These factors suggest that simple physical tests can aid in predicting difficult laryngoscopies, however, more in-depth investigation into these parameters should be performed to draw concrete conclusions.
Ultrasound technology has recently been applied to the airway imaging field in recent years because it is a non-invasive and portable modality. Air and bone are considered to be the two major technical problems involved in ultrasound imaging, however, the artifacts induced by these substances can also be used as important diagnostic tools as long as their causes are understood. For example, ultrasound imaging has been previously explored for predicting difficult airways by detecting the artifactual air signal within the airway structure8,9. With these foundations in place, clinical studies involving the prediction of a difficult airway are becoming more popular. Hui et al.16 suggests that sublingual ultrasound can serve as a potential tool for predicting a difficult airway as a complementary measure to classical prediction methods. Along these lines, some studies suggest that the volume and thickness of the tongue can predict a difficult airway17, 18, whereas other studies have implicated the neck circumference as a major predictor19,29. More related to the current work, some studies have measured the anterior soft tissue thickness via ultrasound for predicting a difficult laryngoscopy8,20,21, but these studies have yet not established a standard for which method is best.
In the current study, the ultrasound probe was placed along the parasagittal plane and mainly focused on the characteristics of the larynx structure itself. For men, the larynx is often much higher compared to women, and poor probe contact in these locations sometimes limited the visualization of the larynx structures and the median sagittal measurements. Prasad et al.22 showed that the epiglottis can be seen in both the anterior transverse cervical plane and in the parasagittal plane; the epiglottis was more distinguishable between the hyoid bone and the thyroid cartilage in the parasagittal view. In current study, the epiglottis can be clearly seen in the parasagittal view, and the DST, DTE and DSE can be accurately and reliably measured (Figure 1).
Pinto et al.23 evaluated the use of the ultrasound-measured distance from the skin to epiglottis in the transverse plane and demonstrated that a cutoff value of 2.75 cm was effective for classifying easy vs difficult laryngoscopies. Falcetta et al.24 also measured this same distance and found that a cutoff value of 2.54 cm was the most effective. Contrary to both of these previous works, we found that a DSE cutoff value of 2.36 cm was optimal, thus further presenting a level of variability that needs to be accounted for and/or corrected in future research. In the parasagittal plane, the DSE is the distance between the hyoid bone and thyroid cartilage (Figure 1 and Figure 2), the ambiguity and movement of which may be why there is no accepted standard. According to the schematic drawing of the sagittal section of the larynx studied by Reidenbach et al.25, we can clearly see the adjoining relationship of the various laryngeal structures. However, the region covered by the hypoepiglottic ligament can greatly change by lifting the epiglottis during intubation. We chose to measure at the upper rim of the thyroid cartilage, partly because of the bony markers in the location, and partly because the pre-epiglottal space is less affected by epiglottis movement during intubation. The DTE we measured incorporated the pre-epiglottal space composed of fat pads. To further analyze whether a difficulty laryngoscopy is related to subcutaneous fat at the upper rim of the thyroid cartilage, we also measured the DST and DSE, which is in fact the sum of the DST and DTE (Figure 1). Our results indicate that the DSE can serve as an independent predictor of a difficult laryngoscopy, but not the DST or DTE. This result suggests that if subcutaneous fat is thick at the level of the thyroid cartilage (DST) or there exists a large pre-epiglottal space (DTE), a difficulty laryngoscopy cannot necessarily be predicted with confidence. However, when both of these features are present, the visual path to explore the glottis is noticeably obstructed and presents a scenario that can much more effectively predict the occurrence of a difficulty laryngoscopy.
Current airway evaluation methods can be predictive but not definitive of difficult intubations, and a noticeable error rate exists because intubation difficulties are inherently subjective. The experience and ability of the anesthesiologist are likely the most important factors of a successful intubation, however, we chose to use the first laryngoscopic view of the CL classification as a replacement indicator of difficult intubation. In the study, the incidence of difficult laryngoscopies at the first view was 20.85%, which was similar to other reports in literature26 -28. Various studies29,30 have shown that men are more likely to have difficulty with the laryngoscopy procedure, and our study also shows that men are at higher risk for a difficulty laryngoscopy. When considering the subject’s BMI, we used the cut-off value of 25, as set by the WHO31, to define overweight. Quinn et al.’s research shows that for every 1-point increase in BMI, there is a 7 % increased risk of intubation failure. The modified Mallampati classification is a commonly used method of airway assessment in the clinic32 and has been shown in previous studies to produce a wide range of sensitivity (42%-81%) and specificity (66%-84%) values for predicting a difficult laryngoscopy33,34: In our study, the MMT displayed a sensitivity of 0.750 (95% CI: 0.692-0.808) and specificity of 0.713 (95% CI: 0.652-0.774), which are consistent with previous studies. Despite these promising results, of all measurements collected in the current study, the DSE was the only one that was found to be a statistically significant independent indicator for predicting difficulty laryngoscopies, resulting in a sensitivity of 0.818 (95% CI: 0.766-0.870) and specificity of 0.856 (95% CI: 0.809-0.904). Nevertheless, by utilizing a “best” model, constructed with other indicators in addition to the DSE, to predict the difficult airway, the AUC reached 93.28%.
Study Limitations
In the current work, we standardized many variables that include the laryngoscopy equipment, experience of anesthesiologist, procedure for the first view of the glottis (used for Cormack-Lehane classification, without external laryngeal maneuvers during classification), but the fact that different physicians subjectively graded the laryngoscope view is a major source of uncertainty. Although these methods were standardized, there is still the possibility of bias and subjectivity due to the individual opinion of each anesthesiologist. Additionally, only one ultrasound machine and one special experienced anesthesiologist performed the ultrasound airway evaluation and it was difficult for us to enroll all patients who met the inclusion criteria. Therefore, a very small number of patients were selected but did not receive a preoperative ultrasound evaluation. Therefore, it was difficult for the subjects to be randomly selected and a bias may have remained. Finally, in two cases, the thyroid cartilage was obviously calcified, and the accurate measurement of the DST and DTE was limited (with no effect on the DSE measurements).
In conclusion, the DSE was found to be an independent predictor of a difficult laryngoscopy. The DSE cutoff value of 2.36 cm resulted in a more powerful predictive value than other indicators for predicting a difficult laryngoscopy. Nevertheless, the combination of various parameters in a “best” model was the ideal case for predicting a difficult laryngoscopy in the current study.