Abstract
Background and Aims:
Effective airway assessment prevents intubation failure. Despite the limited sensitivity of standard tools, ultrasonography (USG) offers promise, especially in predicting difficult laryngoscopies. Our study aims to evaluate tongue-to-oral height ratio (TTOHR) and anterior neck soft tissue measurements on airway USG to predict difficult laryngoscopy.
Methods:
This prospective observational study enroled 120 patients posted for elective surgery under general anaesthesia, without anticipated difficult airways. The skin to hyoid bone distance (SHBD), skin to epiglottis distance (SED), skin to thyrohyoid membrane distance (STHMD), and TTOHR were measured on USG. Modified Cormack-Lehane (CL) grading at laryngoscopy was the primary outcome. Statistical analysis included ROC curve analysis, multivariable logistic regression, and evaluation of predictive models combining multiple USG parameters.
Results:
The incidence of difficult laryngoscopy and intubation was 11.6% and 6.6%, respectively. The highest diagnostic performance was observed for SED, with an area under the curve (AUC) of 0.95 [95% confidence interval (CI): 0.91, 0.98], a cut-off value of 1.87 cm, a sensitivity of 100%, a specificity of 89%, and a diagnostic accuracy of 90%. STHMD followed with an AUC of 0.94 (95% CI: 0.88, 0.99), a cut-off of 1.58 cm, a sensitivity of 90%, a specificity of 86%, and a diagnostic accuracy of 84%. TTOHR showed an AUC of 0.92 (95% CI: 0.78, 1.00), with a cut-off value of 0.80, a sensitivity of 92%, a specificity of 98%, and the highest diagnostic accuracy of 97% (95% CI: 0.96, 1.00). Various models, tested using three or four parameters, showed AUC values ranging from 0.96 to 0.97. A model containing TTOHR, SHBD, and STHMD was identified as a best-fit model for predicting difficult laryngoscopy.
Conclusion:
SED, STHMD, and TTOHR individually showed substantial diagnostic accuracy, with AUCs ranging from 0.92 to 0.95. Analysis of models combining parameters outperformed individual measurements, with statistically significant contributions from TTOHR and SHBD.
Keywords: Airway assessment, airway management, difficult airway, Mallampati score, skin-to-epiglottis distance, skin-to-thyrohyoid membrane distance, tongue height-to-oral height ratio, tongue thickness, ultrasonography
INTRODUCTION
Airway management is a critical component of anaesthetic practice, with difficult laryngoscopy (DL) posing significant challenges and potential risks, including failed intubation and adverse patient outcomes. Notably, 1.5%–13% of patients initially classified as having an easy airway experience unexpected difficulty,[1] highlighting limitations in current assessment techniques. A Cochrane meta-analysis by Roth et al.[2] highlighted the unreliability of bedside clinical screening tests, particularly in patients with apparently normal airway anatomy. This emphasises the need for improved tools to predict unanticipated airway challenges in patients initially deemed easy to manage.[3,4]
Despite the promising nature of ultrasonography (USG)-derived parameters, which have been previously demonstrated with considerable accuracy for various anterior neck soft tissue measurements, including skin-to-hyoid bone distance (SHBD), skin-to-epiglottis distance (SED), and skin-to-thyrohyoid membrane distance (STHMD),[5,6,7] significant gaps remain in the literature.[2] Additionally, methodological inconsistencies, such as differences in probe orientation (sagittal vs coronal) and measurement landmarks, further complicate comparisons and the generalisability of the findings.[8]
This study aims to address this gap and assess the diagnostic accuracy of individual and combined USG parameters in predicting DL in patients with unanticipated difficult airways. The primary objective was to evaluate various upper airway USG parameters from the anterior neck (SHBD, SED, and STHMD) and the newer tongue-to-oral height ratio (TTOHR). Secondary objectives included determining cut-off values for these parameters and analysing combinations to identify the best predictors of DL. We hypothesised that upper airway parameters, specifically the TTOHR alone or in combination with various other parameters, would demonstrate a high diagnostic accuracy in predicting DL in unanticipated difficult airways.
METHODS
This prospective observational study received Institutional Ethics Committee approval (vide approval number: IEC/AIIMS BBSR/2020-21/42, dated 22 June 2020). The study was prospectively registered in the Clinical Trials Registry-India (vide registration number CTRI/2020/08/027487, available at https://ctri.nic.in/, dated 31 August 2020) and recruited patients from September 2020 to February 2022. All patients provided written informed consent for participation in the study and the use of their data for research and educational purposes. The study adhered to the Declaration of Helsinki (2013) and Good Clinical Practice guidelines and followed the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines.
Participants included elective surgical patients aged 18–65 years, scheduled for general anaesthesia and oro-tracheal intubation, with American Society of Anesthesiologists physical status I-III. Exclusions were patients who refused to participate, required fibreoptic intubation, had anticipated difficult airways (Modified Mallampati (MMP) grading ≥ III, mouth opening <3 fingers, Thyromental distance (TMD) <6.5 cm), or had altered neck anatomy unsuitable for USG. Demographic data and clinical screening tests (mouth opening, MMP grading, and TMD) were collected by an unaffiliated anaesthesiologist to assess airway difficulty.
The principal investigator (SN), who had prior experience with over 50 supervised USG examinations, performed airway USG for all participants by using a portable USG machine (SonoSite M Turbo® USG machine, Fujifilm Sonosite, Bothell, Washington, USA) with a linear transducer (6–13 MHz) in the preoperative period. Four USG measurements were obtained for each patient by using transverse and sagittal probe orientations. The SHBD, SED, and TTOHR were measured in a neutral neck position with the transversely oriented probe. The SHBD was recorded from the skin to the hyoid bone, identified as a curved echogenic structure with a posterior acoustic shadow [Figure 1a].[9] Similarly, the SED was measured with the epiglottis identified as a curvilinear hypoechoic structure, characterised by a bright posterior air-mucosal interface (AMI) and a hyperechoic pre-epiglottic space[9] [Figure 1b]. The STHMD was measured by positioning the probe in a sagittal orientation with the neck slightly extended to facilitate probe placement [Figure 1c]. It was identified as a hyperechoic band extending between the hyoid and thyroid, with the distance from the skin measured at the level of the epiglottis.[10] The tongue height was measured from the geniohyoid (GH) [Figure 1d] to the highest point on the dorsum of the tongue, which was identified by the AMI with the hyperechoic palate inferiorly. The oral height was measured from GH to the palate [Figure 1d]. The GH muscles were identified as paired hypoechoic bands inferior to the genioglossus (GG) muscle, which runs in a fan-shaped direction towards the dorsum of the tongue. The investigator (SN) conducting the airway USG was not involved in the patient’s clinical management, including endotracheal intubation (ETI).
Figure 1.
Sonographic landmarks: (a) Skin to hyoid bone distance (SHBD), HB= hyoid bone, (b) Skin to epiglottis distance (SED), EPI= Epiglottis, (c) Skin to thyrohyoid membrane distance (STHMD), THM= thyrohyoid membrane, (d) Tongue height and oral height measurement. GH = geniohyoid; GG = genioglossus, Blue line = oral height, Red line = Tongue height
A modified Cormack-Lehane (CL) grading was noted at laryngoscopy by the intubator, a senior anaesthesiologist with more than 3 years of experience who was not part of the study, and was confirmed by the attending consultant. A modified CL grading of 1 and 2 was defined as easy laryngoscopy (EL), and CL grades 3 and above were defined as DL. Difficult intubation (DI) was defined as requiring more than three attempts or two attempts with the change of operator, use of a higher lifting force, and/or an alternating technique or equipment, such as a bougie, stylet, or video laryngoscope. The first intubation attempt was done using a Macintosh laryngoscope with size 3 or 4 blades. The USG findings were not revealed to the anaesthesiologist performing tracheal intubation.
All USG images were independently reviewed by a senior consultant (BR), who was unaware of the tracheal intubation characteristics and not involved in patient management. Discrepancies in interpretation led to the exclusion of corresponding patient data from the final analysis.
The sample size calculation was performed using RStudio Desktop version 4.2.2 with the pROC package for diagnostic tests.[11] A previous study reported that the incidence of DL in Indian patients with apparently normal airways was 9.7%.[10] Assuming that the USG parameters have a good discriminatory ability [defined as a minimum area under the curve (AUC) = 0.75] to detect DL and considering an event rate of 10% (meaning that 10% of patients will experience DL), we required a sample of 120 patients to be reasonably certain (80% power, 5% alpha) in the ratio of 1:10 (cases to controls). Due to the 10% local prevalence of the condition, the sample size was calculated with a 1:10 ratio (cases to controls), resulting in 11 cases and 109 controls.
Data were collected and maintained on Microsoft Excel, and analysis was performed using RStudio Desktop version 4.2.2 and the pROC package.[11] The normality of distribution was assessed using the Shapiro-Wilk test. All data were normally distributed. Continuous data, such as age, weight, height, body mass index (BMI), and USG measurements, were represented as mean and standard deviation. Categorical data, such as gender, ASA classification, and MMP classification, were expressed as proportions and percentages. Students’ t-tests were applied to analyse quantitative data between the EL and DL groups. Receiver operating characteristics (ROC) curves were used to determine the AUC, sensitivity, specificity, and the positive and negative predictive values of the USG measurements, and exploratory cut-offs were computed for each measure to achieve the best possible sensitivity, specificity, and positive and negative predictive values. Differences were considered statistically significant at P < 0.05. The models were constructed by combining three or four parameters, and logistic regression analysis was done after data normalisation on the R platform by using the min-max method.
RESULTS
A total of 163 patients were screened for eligibility, out of which 13 were excluded due to refusal to consent. A total of 120 patients, aged between 18 and 65 years, were included in this study for final analysis [Figure 2].
Figure 2.
Flow chart of study as per STROBE statement
Table 1 summarises the demographic characteristics of the study population. The groups classified as ‘easy’ and ‘difficult’ laryngoscopy were comparable across most variables, except for BMI, weight, and MMP grading, which showed statistically significant differences.
Table 1.
Baseline characteristics of the study population
| Variables | Difficult Laryngoscopy (n=14) | Easy Laryngoscopy (n=106) | P |
|---|---|---|---|
| Gender | |||
| Male | 7 (50%) | 51 (48%) | 0.826 |
| Female | 7 (50%) | 55 (52%) | |
| Age (years) | 47.8 (9.2) (43.7, 51.9) |
41.3 (12.9) (8.8, 43.8) |
0.068 |
| Height (m) | 1.63 (0.05) (1.60, 1.66) |
1.63 (0.04) (1.62, 1.64) |
0.698 |
| Weight (kg) | 65.34 (12.56) (58.4, 72) |
59.2 (10.2) (58, 61.9) |
0.007 |
| BMI (kg/m2) | 24.5 (3.6) (22.4, 26.5) |
22.5 (3.2) (21.7, 23) |
0.023 |
| ASA physical status | |||
| 1 | 4 (3%) | 62 (52%) | 0.06 |
| 2 | 10 (8%) | 44 (36%) | |
| MMP score | |||
| 1 | 2 (1.67%) | 38 (31.6%) | 0.013 |
| 2 | 12 (10%) | 68 (56.7%) |
Data expressed as mean (standard deviation) (95% confidence intervals) and numbers (percentage). n=number of patients; BMI=body mass index; ASA=American Society of Anesthesiologists; MMP=Modified Mallampati
Out of 120 patients enroled, 106 (88.3%) were classified as having an ‘EL’ based on the Modified CL grading, with 66 patients graded as CL 1, 25 as CL 2a, and 15 as CL 2b. The remaining 14 patients (11.6%) were classified under ‘DL’, comprising nine patients with CL 3a, three with CL 3b, and two with CL 4. DI was seen in eight out of 120 patients (6.6%). All DIs were managed either with the assistance of a bougie, a video laryngoscope, or by transferring the case to a more experienced operator.
The means of all USG parameters were significantly greater in patients with DL and DI compared to those with EL and easy intubation [Table 2]. ROC: For every USG parameter (SHBD, STHMD, SED, and TTOHR), ROCs were drawn, and the data are represented in Table 3 and Figure 3. Among the USG parameters, SED showed the highest diagnostic performance with an AUC of 0.95 (95% CI: 0.91, 0.98), followed by STHMD at 0.94 (95% CI: 0.88, 0.99) and TTOHR at 0.92 95 [95% confidence interval (CI): 0.78, 1.00]. TTOHR demonstrated the highest diagnostic accuracy (97%) and the strongest positive likelihood ratio (46), highlighting its superior predictive value. While SHBD had an AUC of 0.85 (95% CI: 0.78, 0.93) with 100% sensitivity, it showed lower specificity (66%) and the lowest positive predictive value (PPV) (28%).
Table 2.
Ultrasonographic parameters in patients with easy and difficult laryngoscopy and intubation
| USG parameters | Easy Laryngoscopy (n=106) | Difficult Laryngoscopy (n=14) | P | Easy Intubation (n=112) | Difficult Intubation (n=8) | P |
|---|---|---|---|---|---|---|
| SHBD (cm) | 0.62 (0.20) (0.63, 0.70) | 0.93 (0.16) (0.83, 1.02) | <0.001 | 0.67 0.20) (0.64, 0.72) | 0.98 (0.13) (0.87, 1.09) | <0.001 |
| STHMD (cm) | 1.41 (0.15) (1.38, 1.44) | 1.70 (0.09) (1.64, 1.75) | <0.001 | 1.42 (0.16) (1.39, 1.45) | 1.71 (0.07) (1.66, 1.78) | <0.001 |
| SED (cm) | 1.67 (0.19) (1.64, 1.71) | 2.03 (0.13) (1.96, 2.11) | <0.001 | 1.69 (0.20) (1.65, 1.73) | 2.06 (0.12) (1.96, 2.16) | <0.001 |
| TTOHR (ratio) | 0.75 (0.03) (0.75, 0.76) | 0.82 (0.08) (0.78, 0.86) | <0.001 | 0.76 (0.03) (076, 0.77) | 0.81 (0.10) (0.72, 0.89) | 0.004 |
Data expressed as mean (standard deviation) (95% confidence interval). n=number of patients; USG=ultrasonography; SHBD=skin to hyoid bone distance; SED=skin to epiglottis distance; STHMD=skin to thyrohyoid membrane distance; TTOHR=tongue to oral height ratio
Table 3.
Receiver operative characteristics of USG parameters for difficult laryngoscopy
| USG Parameters | AUC Mean (95% CI) | Cut-off value (Mean) | Sensitivity Mean (95% CI) | Specificity Mean (95% CI) | Positive predictive value Mean | Negative predictive value Mean | Diagnostic accuracy Mean (95% CI) | Positive LR Mean | Negative LR Mean |
|---|---|---|---|---|---|---|---|---|---|
| SHBD (cm) | 0.85 (0.78, 0.93) | 0.69 | 1.00 | 0.66 (0.54, 0.72) | 0.28 | 1.0 | 0.7 (0.6, 0.77) | 2.94 | 0 |
| STHMD (cm) | 0.94 (0.88, 0.99) | 1.58 | 0.90 (0.64, 1) | 0.86 (0.76, 0.90) | 0.54 | 0.97 | 0.84 (0.77, 0.9) | 6.42 | 0.1 |
| SED (cm) | 0.95 (0.91, 0.98) | 1.87 | 1.00 | 0.89 (0.82, 0.94) | 0.56 | 1.0 | 0.90 (0.84, 0.95) | 9.09 | 0 |
| TH (cm) | 0.88 (0.74, 1.00) | 2.6 | 0.93 (0.78, 1.0) | 0.83 (0.74, 0.88) | 0.56 | 0.97 | 0.83 (0.76, 0.9 | 5.47 | 0.08 |
| TTOHR (ratio) | 0.92 (0.78, 1.00) | 0.81 | 0.92 (0.78, 1) | 0.98 (0.95, 1) | 0.87 | 0.99 | 0.97 (0.94, 1.0) | 46 | 0.08 |
Data expressed as mean or mean (95% confidence interval). The cut-off value is represented as an absolute value, and the rest are represented as ratios. USG=ultrasonography; SHBD=skin to hyoid bone distance; STHMD=skin to thyrohyoid membrane distance; SED=skin to epiglottis distance; TH=tongue height; TTOHR=tongue to oral height ratio; AUC=area under the curve; CI=confidence interval; LR=likelihood ratio
Figure 3.
Receiver operating characteristics analysis of ultrasound parameters. Annotations: Purple: Tongue to oral height ratio (TTOHR); Golden: Skin to epiglottis distance (SED); Green: Skin to thyrohyoid membrane distance (STHMD); Blue: Skin to hyoid bone distance (SHBD)
A binomial multivariate logistic regression analysis was conducted to assess the effects of various demographic parameters, including age, height, weight, BMI, ASA, MPG score, and individual USG parameters, on the difficulty of laryngoscopy. The analysis revealed that the USG parameters, particularly TTOHR and SHBD, remained statistically significant and independent predictors of DL [Table 4].
Table 4.
Multivariate logistic regression analysis of ultrasonographic and clinical predictors of difficult laryngoscopy
| Predictor | Estimate | SE | Z | P |
|---|---|---|---|---|
| Intercept | 249.073 | 203.054 | 1.227 | 0.220 |
| Age | 0.301 | 0.166 | 1.811 | 0.070 |
| Height | −2.202 | 1.437 | −1.532 | 0.125 |
| Weight | 2.538 | 1.654 | 1.535 | 0.125 |
| BMI | −7.327 | 4.599 | −1.593 | 0.111 |
| ASA Grade | −2.760 | 2.359 | −1.170 | 0.242 |
| MPG Score | 0.748 | 1.647 | 0.454 | 0.650 |
| SHBD | 15.660 | 7.852 | 1.994 | 0.046 |
| STHMD | 14.106 | 10.713 | 1.317 | 0.188 |
| SED | 15.307 | 9.533 | 1.606 | 0.108 |
| TTOHR | 56.306 | 27.611 | 2.039 | 0.041 |
ASA=American Society of Anesthesiologists; MPG=Mallampati grading; SHBD=skin to hyoid bone distance; STHMD=skin to thyromental membrane distance; SED=skin to epiglottis distance; TTOHR=tongue to oral height ratio
The validity of five models, based on various combinations of USG-based parameters, was also assessed through multiple logistic regression-derived ROC analysis after data normalisation on the R platform [Table 5, Figure 4]. Normalisation of data was required, as TTOHR is measured as a ratio, whereas the rest are measured as distances. The AUC for combinations of parameters was higher than that of individual parameters. All models containing TTOHR showed an AUC of 0.97. TTOHR and SHBD carried significant weightage, as indicated by OR in three out of four models. Based on Akaike’s Information Criterion for model selection, we found that model 3, which contains the TTOHR, STHMD, and SHBD, is the best-fit model for predicting DL.
Table 5.
Diagnostic validity of various models combining USG parameters for predicting difficult laryngoscopy
| Variable | USG parameters | Log OR (95% CI) | P | AUC | P |
|---|---|---|---|---|---|
| Model 1 | TTOHR STHMD SHBD SED |
6.1*(0.5, 11.7) 6.8 (−3.2, 16.8) 4.9* (0.1, 9.7) 6.7 (−3.7, 7.4) |
0.03 0.18 0.04 0.19 |
0.972 | 0.0001 |
| Model 2 | TTOHR STHMD SED |
5.9 (−0.3, 12.1) 9.4 (−1.2, 20) 3.7 (−5.9, 13.3) |
0.05 0.07 0.44 |
0.972 | 0.0004 |
| Model 3 | TTOHR STHMD SHBD |
5.6* (0, 11.2) 11.9* (4.7, 19.1) 4.1 (−0.3, 8.5) |
0.04 0.01 0.06 |
0.972 | 0.0004 |
| Model 4 | TTOHR SHBD SED |
6.7* (0.7, 12.7) 5.5* (1, 9.9) 11.6* (4.4, 18.8) |
0.03 0.02 0.01 |
0.970 | 0.0001 |
| Model 5 | STHMD SHBD SED |
8.7 (−1.1, 18.5) 4.2* (0, 8.4) 4.1 (−5.5, 13.7) |
0.08 0.04 0.39 |
0.958 | 0.0004 |
*Carries significant weightage, data are expressed as Log OR (lower and upper limits of 95% CI=confidence intervals). USG=ultrasonography; SHBD=skin to hyoid bone distance; STHMD=skin to thyrohyoid membrane; SED=skin to epiglottis; TTOHR=tongue to oral height ratio; SED=skin to epiglottis
Figure 4.
Receiver operating characteristics analysis of various models. TTOHR = tongue to oral height ratio; SHBD = skin to hyoid bone distance; STHMD = skin to thyrohyoid membrane distance; SED = skin to epiglottis distance
DISCUSSION
This study confirms the diagnostic value of USG airway parameters—individually and in combination—measured through the anterior neck and submandibular regions in predicting DL. Prior studies have identified three sonographic domains relevant to airway assessment: anterior neck soft tissue thickness, anatomic positioning (APD), and oral space domain (OSD).[8] Bhargava et al.[8] reported APD parameters, particularly hyomental distance (HMD), as the most accurate predictors. Building on this, Lin et al.[12] proposed the DARES protocol, incorporating key elements from all three domains, including SED, tongue height, and HMD. Our study reinforces and extends this framework by demonstrating that integrating OSD—specifically, the novel TTOHR—significantly enhances predictive accuracy.
Previous studies have reported inconsistent predictive value for tongue-related parameters. While Ohri et al.[13] found high sensitivity (95%) for tongue volume, others, such as Parameswari et al.,[14] showed moderate accuracy, and Ning et al.[15] reported no significant association, likely due to tongue width having a minimal impact on laryngoscopy. Another recent meta-analysis reported tongue height as a more reliable predictor than volume, given the static nature of tongue width during laryngoscopy.[4]
In our study, the tongue height had a lower cut-off value (2.6 cm) compared to prior reports, such as those by Yao et al.[16] (>6.1 cm) and Agarwal et al.[17] (>5.8 cm). Yet, it demonstrated good predictive value for DI. This may be due to our use of the geniohyoid-to-dorsum distance rather than the skin-to-dorsum distance, and our exclusion of patients with a MMP grade of ≥3, which focuses on unanticipated difficult airways.
TTOHR, a novel OSD parameter, was previously studied by Andruszkiewicz et al.,[18] but with differing results, likely due to their sagittal probe orientation and inclusion of patients with restricted mouth opening. Our coronal approach, aligned with Bhargava et al.’s[8] recommendations, and standardised patient selection may explain our more favourable findings.
Among the anterior neck soft tissue parameters, SED emerged as the top predictor for DL in our cohort, with an AUC of 0.95, a sensitivity of 100%, a specificity of 89%, and a diagnostic accuracy of 90%, using a cut-off of 1.87 cm. In their meta-analysis, Carsetti et al.[7] reported a similar sensitivity of approximately 82% for SED. Shetty et al.[19] and Karimbanakkal et al.[20] found a cut-off value similar to ours in a South-Indian population. Other studies reported cut-off values ranging from up to 2.8 cm, possibly reflecting ethnic variations.[21,22]
STHMD emerged as the next reliable predictor after SED. In contrast to our study, which used sagittal probe placement, previous studies with a transverse orientation of the probe found higher cut-off values.[10,23] A recent study similarly identified a cut-off of 1.6 cm for STHMD when measured using a sagittal probe orientation, similar to ours.[21]
The SHBD cut-off value of 0.69 cm predicted DL with 100% sensitivity. Yadav et al.[10] also reported a mean value of 0.7 cm in DL, while Archana et al.[23] reported a cut-off of 1.26 cm to be predictive of DL with a sensitivity of 95%.
Given the low incidence of DL in patients with unanticipated difficult airways, tests with a higher PPV are crucial for clinical decision-making. In our study, the TTOHR demonstrated an exceptionally high PPV, establishing it as a key parameter for predicting DL. Furthermore, we evaluated the performance of five distinct models incorporating various combinations of USG parameters. While all the models demonstrated superior validity against the individual parameters, three models notably showed higher diagnostic accuracies with TTOHR as the common parameter in all three. These combinations reinforce the clinical value of TTOHR and highlight the potential of multimodal USG assessments to improve the accuracy of airway prediction, particularly in unanticipated challenging airway scenarios.
Our study has several limitations. Firstly, we studied only a small population from a particular ethnicity, which may not always be applicable to other ethnicities. Additionally, these values should not be considered for obese, pregnant, and other anticipated populations with difficult airways. Secondly, the tongue is a mobile structure; measurements may reveal variations in its shape. To address this limitation, we recommend assessing it compared to oral cavity height. Our study findings do not evaluate the accuracy of these tests in predicting DI, as a larger sample size would be needed to assess their role in such predictions. Lastly, the measurements were taken by a single investigator and verified by a second investigator. We cannot validate the intra- and inter-rater variability.
CONCLUSION
Our study demonstrated high diagnostic accuracy for all parameters derived from the anterior neck and oral space in predicting difficult laryngoscopy in patients with unanticipated difficult airways, with skin-to-epiglottis distance identified as the single best predictor based on area under the curve and sensitivity, and tongue-to-oral height ratio to have the highest diagnostic accuracy and positive predictive value. Among the tested models that combined various parameters, the model incorporating the tongue-to-oral height ratio, skin-to-thyrohyoid membrane distance, and skin-to-hyoid bone distance emerged as the best fit.
Authors Contributions
SN: Concepts, Design, Investigation, Data Collection, Data Analysis, Manuscript Writing, Editing; AP: Concepts, Design, Investigation, Manuscript Writing, Editing. NM: Concepts, Design, Investigation, Data Analysis, Manuscript Writing, Editing; PBR: Investigation, Data Collection, Data Analysis, Manuscript Writing and Editing; AS: Data Analysis and Manuscript Editing.
Study data availability
Study data availability: De-identified data may be requested with reasonable justification from the authors (email to the corresponding author) and will be shared after approval, as per the authors’ Institution’s policy.
Disclosure of use of artificial intelligence (AI)-assistive or generative tools
The AI tools or language models (LLM) have not been utilised in the manuscript, except that software has been used for grammar corrections and reference.
Presentation at conferences/CMEs and abstract publication
Nil.
Conflicts of interest
There are no conflicts of interest.
Funding Statement
Nil.
REFERENCES
- 1.Apfelbaum JL, Hagberg CA, Caplan RA, Blitt CD, Connis RT, Nickinovich DG, et al. Practice guidelines for management of the difficult airway: An updated report by the American Society of Anesthesiologists Task Force on Management of the Difficult Airway. Anesthesiology. 2013;118:251–70. doi: 10.1097/ALN.0b013e31827773b2. [DOI] [PubMed] [Google Scholar]
- 2.Roth D, Pace NL, Lee A, Hovhannisyan K, Warenits AM, Arrich J, et al. Airway physical examination tests for detection of difficult airway management in apparently normal adult patients. Cochrane Database Syst Rev. 2018;5:CD008874. doi: 10.1002/14651858.CD008874.pub2. doi: 10.1002/14651858.CD008874.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kollmeier BR, Boyette LC editors, editors. StatPearls Publishing; 2025. Difficult Airway. Available from: https://www.ncbi.nlm.nih.gov/books/NBK470224/ . [Last accessed on 2025 Jun 11] [PubMed] [Google Scholar]
- 4.Gomes SH, Simões AM, Nunes AM, Pereira MV, Teoh WH, Costa PS, et al. Useful ultrasonographic parameters to predict difficult laryngoscopy and difficult tracheal intubation-A systematic review and meta-analysis. Front Med (Lausanne) 2021;28:671658. doi: 10.3389/fmed.2021.671658. doi: 10.3389/fmed.2021.671658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lazar AE, Gherghinescu MC. Airway ultrasound for anesthesia and in intensive care patients-A narrative review of the literature. J Clin Med. 2022;11:6327. doi: 10.3390/jcm11216327. doi: 10.3390/jcm11216327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Petrisor C, Dîrzu D, Trancă S, Hagău N, Bodolea C. Preoperative difficult airway prediction using suprahyoid and infrahyoid ultrasonography derived measurements in anesthesiology. J Med Ultrason. 2019;21:83–8. doi: 10.11152/mu-1764. [DOI] [PubMed] [Google Scholar]
- 7.Carsetti A, Sorbello M, Adrario E, Donati A, Falcetta S. Airway ultrasound as predictor of difficult direct laryngoscopy: A systematic review and meta-analysis. Anesth Analg. 2022;134:740–50. doi: 10.1213/ANE.0000000000005839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bhargava V, Rockwell NA, Tawfik D, Haileselassie B, Petrisor C, Su E. Prediction of difficult laryngoscopy using ultrasound: A systematic review and meta-analysis. Crit Care Med. 2023;51:117–26. doi: 10.1097/CCM.0000000000005711. [DOI] [PubMed] [Google Scholar]
- 9.Kundra P, Mishra SK, Ramesh A. Ultrasound of the airway. Indian J Anaesth. 2011;55:456–62. doi: 10.4103/0019-5049.89868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Yadav NK, Rudingwa P, Mishra S, Pannerselvam S. Ultrasound measurement of anterior neck soft tissue and tongue thickness to predict difficult laryngoscopy - An observational analytical study. Indian J Anaesth. 2019;63:629. doi: 10.4103/ija.IJA_270_19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez J-C, et al. pROC: An open-source package for R and S+to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77. doi: 10.1186/1471-2105-12-77. doi: 10.1186/1471-2105-12-77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lin J, Bellinger R, Shedd A, Wolfshohl J, Walker J, Healy J, et al. Point-of-care ultrasound in airway evaluation and management: A comprehensive review. Diagnostics (Basel) 2023;13:1541. doi: 10.3390/diagnostics13091541. doi: 10.3390/diagnostics13091541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ohri R, Malhotra K. Different 2D ultrasound calculation methods to evaluate tongue volume for prediction of difficult laryngoscopy. Indian J Anaesth. 2020;64:S193–7. doi: 10.4103/ija.IJA_843_19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Parameswari A, Govind M, Vakamudi M. Correlation between preoperative ultrasonographic airway assessment and laryngoscopic view in adult patients: A prospective study. J Anaesthesiol Clin Pharmacol. 2017;33:353–8. doi: 10.4103/joacp.JOACP_166_17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ning L, Zhu X, Li H-C, Zhou S-J, Zhang Q-W, Zou H-Y, et al. A quantitative study of airway ultrasound in predicting difficult laryngoscopy: A prospective study. Chin J Traumatol. 2023;26:351–6. doi: 10.1016/j.cjtee.2023.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yao W, Wang B. Can tongue thickness measured by ultrasonography predict difficult tracheal intubation? Br J Anaesth. 2017;118:601–9. doi: 10.1093/bja/aex051. [DOI] [PubMed] [Google Scholar]
- 17.Agarwal R, Jain G, Agarwal A, Govil N. Effectiveness of four ultrasonographic parameters as predictors of difficult intubation in patients without anticipated difficult airway. Korean J Anesthesiol. 2021;74:134–41. doi: 10.4097/kja.20114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Andruszkiewicz P, Wojtczak J, Sobczyk D, Stach O, Kowalik I. Effectiveness and validity of sonographic upper airway evaluation to predict difficult laryngoscopy. J Ultrasound Med. 2016;35:2243–52. doi: 10.7863/ultra.15.11098. [DOI] [PubMed] [Google Scholar]
- 19.Shetty SR, V.T S. Validation of clinical versus ultrasound parameters in assessment of airway. Trends in Anaesth and Crit Care. 2020;35:21–7. [Google Scholar]
- 20.Karimbanakkal F, Rajesh MC, Bau SP. Comparison between El-Ganzouri risk index and sonologically measured skin-to-epiglottis distance for predicting difficult laryngoscopy: A prospective observational study. Indian J Anaesth. 2025;69:380–5. doi: 10.4103/ija.ija_896_24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.De Luis-Cabezón N, Ly-Liu D, Renedo-Corcostegui P, Santaolalla-Montoya F, Zabala-Lopez de Maturana A, Herrero-Herrero JC, et al. A new score for airway assessment using clinical and ultrasound parameters. Front Med (Lausanne) 2024;11:1334595. doi: 10.3389/fmed.2024.1334595. doi: 10.3389/fmed. 2024.1334595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Falcetta S, Cavallo S, Gabbanelli V, Pelaia P, Sorbello M, Zdravkovic I, et al. Evaluation of two neck ultrasound measurements as predictors of difficult direct laryngoscopy: A prospective observational study. Eur J Anaesth. 2018;35:605–12. doi: 10.1097/EJA.0000000000000832. [DOI] [PubMed] [Google Scholar]
- 23.Archana KN, Ajnas AP, Kumararadhya GB. Ultrasound evaluation of anterior neck soft-tissue thickness to predict difficult intubation in overweight adult patients posted for surgery under general endotracheal anesthesia – An observational study. Ann Afr Med. 2024;23:182–8. doi: 10.4103/aam.aam_33_23. [DOI] [PMC free article] [PubMed] [Google Scholar]