PREDICTION OF DIFFICULT LARYNGOSCOPY BY ULTRASOUND GUIDED VALUATION OF ANTERIOR NECK SOFT TISSUE THICKNESS.

difficult laryngoscopy. However, USG guided measurements at the level of hyoid bone and anterior commissure of vocal cords showed a higher specificity and sensitivity for the prediction of difficult laryngoscopy. With reference to ROC analysis, the optimal cut-offs of DSHB, DSAC, neck circumference and BMI measurements for the prediction of difficult Laryngoscopy is 0.81 cm, 0.92 cm, 35.75cm and 24.8 kg/m2 respectively with the area under the curves being 0.944, 0.970, 0.801 and 0.745 respectively. With reference to ROC analysis, the optimal cut-off value for modified Mallampati grades for the prediction of difficult Laryngoscopy is Grade II and above with area under the curves being 0.718.

The receiver operating characteristic (ROC) curve, which is defined as a plot of test sensitivity as the y coordinate versus its 1-specificity or false positive rate (FPR) as the x coordinate, is an effective method of evaluating the quality or performance of diagnostic tests against a specific Gold standard. This technique was used to find the best predictor of difficult Laryngoscopy.

Observations And Results:-
The study entitled "Prediction of Difficult Laryngoscopy by Ultrasound guided valuation Of Anterior Neck Soft Tissue Thickness" was conducted in 100 patients of both gender and following observations and results were obtained. The following parameters were studied: Modified Mallampati classification (MMC) based on relations of oropharyngeal structures. 1. Thyromental distance 2. Neck circumference 3. USG measurements: 4. Distance from the skin to hyoid bone (DSHB) 5. Distance from the skin to the anterior commissure of vocal cords ( DSAC) 6. Modified Cormack Lehane score (MCLS)     P-value by independent sample t-test. P-value<0.05 is considered to be statistically significant. NS-Statistically nonsignificant. The distribution of mean age did not differ significantly between a group of cases with easy and difficult laryngoscopy (P-value>0.05).  The sex distribution of cases studied did not differ significantly between a group of cases with easy and difficult laryngoscopy (P-value>0.05).  0.001 *** P-value by independent sample t-test. P-value<0.05 is considered to be statistically significant. ***P-value<0.001 (Statistically significant). The distribution of mean BMI is significantly higher among the group of cases with difficult laryngoscopy compared to the group of cases with easy laryngoscopy (P-value<0.001).  .05 is considered to be statistically significant. ***P-value<0.001 (Statistically significant). The distribution of Mallampati grades of cases studied differs significantly between a group of cases with easy and difficult laryngoscopy (P-value<0.001).  P-value by independent sample t-test. P-value<0.05 is considered to be statistically significant. ***P-value<0.001.
The distribution of mean neck circumference is significantly higher among the group of cases with difficult laryngoscopy compared to the group of cases with easy laryngoscopy (P-value<0.001).
Graph 6:-Distribution of mean neck circumference with respect Cormack Lehane grades (MCLS). The distribution of mean thyromental distance did not differ significantly between a group of cases with easy and difficult laryngoscopy (P-value>0.05).  P-values by independent sample t test. P-value<0.05 is considered to be statistically significant. ***P-value<0.001 (Statistically significant).
The distribution of mean DSHB is significantly higher in Difficult Laryngoscopy group compared to easy laryngoscopy group (P-value<0.001).
Graph 8:-Distribution of mean of distance between skin to hyoid bone (DSHB) in relation to the modified Cormack Lehane score.   P-values by independent sample t-test. P-value<0.05 is considered to be statistically significant. ***P-value<0.001 (Statistically significant).
The mean ± SD of time taken to complete the measurement in Group 1 and Group 2 was 1.52 ± 0.54 mins and 1.05 ± 0.13 mins respectively. The distribution of the mean time taken to complete the measurements is significantly higher in Group 1 compared to Group 2 (P-value<0.001). With latest advancements of ultrasonography, attempts have been made to widen its horizon of utility due to its accuracy along with patient-friendly non-invasive technique. In this study, we aimed to assess and prepare beforehand for unanticipated difficult airway in the operation theater.
Our study was a prospective observational type of study, carried out in attached hospital to assess the efficacy of ultrasonogram as a tool to measure the thickness of anterior neck soft tissue thickness at the level of hyoid bone and anterior commissure of vocal cord and correlate with modified Cormack Lehane score to predict difficult laryngoscopy. The patient's demographic profile was analyzed for any statistical significance.

Table 2/ Graph 2:-
This study included patients falling in the age group of 18 -65 yrs of age. The larynx starts to develop as early as the third week of gestation and is completely developed by the end of the second trimester. At infancy, the larynx is present at the level of C2-C3vertebra, with the advancement of age the larynx descends down to the adult level which is C3-C6 vertebra. This descends varies in gender. In females, the descend stops at an early age whereas it stops at puberty in the male population. Whereas with age there are changes in the body which make the airway difficult. Therefore, to prevent any differences in airway measurements due to age-related discrepancies we had excluded patients younger than 18 years or older than 65 years. But age did not differ significantly with respect to MCLS in our study.

Table 3/ Graph 3:-
The sex distribution of cases studied did not differ significantly between a group of cases with easy and difficult laryngoscopy (P-value>0.05).  7 and Wu J et al. 8 reported that difficult intubation increases with increasing Body Mass Index (BMI).Hence our results regarding the BMI of the subjects also support their finding, as the distribution of mean BMI in our study is significantly higher among the group of cases with difficult laryngoscopy compared to the group of cases with easy laryngoscopy (P-value<0.001) evident from table 4 and graph 4.

Table 5/ Graph 5:-
Ezri, et al. 9 , suggested in their study that Mallampati score poorly predicted difficult laryngoscopy.Wu J, et al. 8 , they found the Modified Mallampati score to have a significant p-value <0.0001 with a sensitivity of 50% and specificity of 82.3% concluding that modified Mallampati score was a good parameter in favor to assess difficult laryngoscopy but can be used as a part of multifactorial model for the prediction of difficult tracheal intubation.
Reddy, et al. 10 , also found in their study that modified Mallampati score (≥ 3) is one of the good predictors of difficult intubation with a sensitivity of 71.4 %, specificity of 83.7% Whereas our study concludes with the distribution of Mallampati grades of cases studied differs significantly between a group of cases with easy and difficult laryngoscopy (P-value<0.001); with a sensitivity of 81.2% and specificity 60.7%; but it's accuracy was merely 64%. This shows that the predictability of difficult laryngoscopy with the modified Mallampati score is not a dependable variable.  10 all showed a similar opinion regarding TMD in their study, stating that TMD is not a reliable screening test for predicting difficult laryngoscopy. Results from our study with regards to TMD favor their opinion as a poor predictor for difficult laryngoscopy (evident from Table7 and Graph 7). The measurements taken down with the help of USG were used to identify and label the patient as the difficult airway.  9 in their study arrived with the result that patients with a larger distance between skin and hyoid bone ( >2.8cm ) had difficult laryngoscopy; which was comparable to results from our study. In our study, anterior neck soft tissue thickness at the level of hyoid bone >0.81cm were good predictors of difficult laryngoscopy In 2011, SrikarAdhikari, et al. 12 demonstrated in their study that the sonographic measurement of the anterior neck soft tissue thickness at the level of hyoid bone >1.69cm ( 95 %, CI = 1.19 to 2.19) could be used to distinguish difficult laryngoscopy; this result could be related to our result in which it was evident that distance from skin to hyoid bone of more than 0.81cm was the cut off value to label the patient with difficult laryngoscopy.
Mirulnalini, et al. in 2015 13 , conducted a similar study, in which they found the distance between skin to hyoid bone to be statistically highly significant, with cut off value of 0.78 cm which is close to our result of statistical significance and cut off value of 0.81cm for prediction of difficult laryngoscopy. In contrast to their sensitivity of 63.6%, this parameter had the sensitivity of 93.7%.  10 came up with the result that anterior neck soft tissue thickness at the level of vocal cords of > 0.23cm had a sensitivity of 85.7% in predicting a Cormack Lehane Grade III or IV, which was higher than that of modified Mallampati Grade of 3 or 4 denoting difficult laryngoscopy. Whereas in our study anterior neck soft tissue thickness at the level of vocal cords more than 0.92cm was exclusive for difficult laryngoscopy with the sensitivity of around 97%, comparatively higher than modified mallampati scoring. Though our measurements were greater than the study done by Reddy, et al. our study's inference matched with their conclusion.
Results of a study conducted by Wu J, et al. in 2014 8 , which showed that the optimal cut off values for modified mallampati score, distance from skin to hyoid bone and anterior commissure to predict difficult laryngoscopy were over MMC grade 2, 1.28 cm and 1.1 cm respectively which were comparable to our results; in which the cut off values for modified mallampati score (more than 2), hyoid bone depth (>0.81cm), and anterior commissure dept (0.92cm) were individually associated with difficult laryngoscopy (summarized in table10).
To summarize the results from our study; out of the 16 patients with difficult laryngoscopy, 11 patients were of grade I and II. This means that modified mallampati grading could not predict difficult laryngoscopy in these 11 patients which shows its lower predictability value than the USG measurements. Graph 5 and Table 5

ROC curve analysis of BMI:-
The cutoff point is 24.8 kg/m 2 .Area under the curve is 0.745, which means that accuracy of the test is "fair" Sensitivity was 68 .7% sensitive.Specificity 76.2% was specific.
So based on outcome analysis it was found that ultrasonogram can be used as a safe, less time consuming and potential tool in predicting difficult intubation preoperatively by measuring the thickness of anterior neck soft tissues. The increased thickness of the soft tissues was associated with increased difficulty in intubation.
Also based on ROC analysis, it was evident that USG measurements at the level of anterior commissure ( with range 0.73cm to 1.78cm ) are more accurate, more sensitive and more specific in detecting the occurrence of the difficult intubation followed by measurement at the level of hyoid bone (with range 0.90cm -1.69cm), neck circumference (>35.75CM) and modified mallampati grade (>GRADE 2).

Table 12/ Graph 12
Secondarily, we had statistically analyzed the time required to assess and measure the airway structures using USG. We found that with experience the required time to measure the distances using USG was reduced. Hence based on statistics we can conclude that Ultrasonogram can be used as an independent tool for assessing airway, which will be helpful in cases of emergency intubations without consuming much of the precious time.

Summary:-
We have conducted a prospective observational study to predict difficult laryngoscopy by Ultrasound guided measurement of Anterior Neck Soft Tissue Thickness preoperatively and correlation of these measurements with Cormack-Lehane in 100 patients planned for elective surgery under general anaesthesia requiring intubation.
Our study included demographic details like age and sex, which were comparable. The comparison of BMI with MCLS grading was statistically significant with a higher BMI corresponding with a higher MCLS grade.The modified Mallampati Grade of 2 or more correlated with difficult laryngoscopy in our study. Also, our study showed that the neck circumference of more than 35.75cm indicated difficult intubation. However, the thyromental distance was not statistically significant when compared with MCLS.

255
Whenthese patients were evaluated using USG, the distance from skin to hyoid bone and skin to the anterior commissure of vocal cords were statistically significant when compared with MCLS. We also found that with experience the required time to measure the distances using USG was reduced.

Conclusion:-
We conclude from our study that the BMI, modified Mallampati grade and neck circumference are good predictors of difficult laryngoscopy. However, USG guided measurements at the level of hyoid bone and anterior commissure of vocal cords showed a higher specificity and sensitivity for the prediction of difficult laryngoscopy.
Thus, with advancements in hospital economy and easy availability of portable Ultrasound machines; we as anaesthesiologists can use USG as a clinical tool for assessing airway in order to rule out difficult airway and prepare the anaesthesia workstation for the benefit of the patient. Hence avoid the chaotic scenario of cannot ventilate and cannot intubate.