Airway Ultrasonography Measurements With Cormack Lehane Classification in Patients With OSA

Research design

1. This is a cross sectional study. Involving face-to-face interview, USG airway assessment and Cormack Lehane score assessment. The patient will be identified from locations listed below:

• Anaesthesia clinic
• In the ward, during review by anaesthesia team for elective operation in the following day
• In the ward, when cases are notified for operations as emergency case

1. Once consent obtained from the patient, information regarding surgery and date of surgery will be recorded in the patient's data collection sheet. Patient's STOPBANG score and AHI score will also be recorded.
2. The airway ultrasonography will be performed by the principal investigator using a standardized ultrasound machine (NextGen LOGIC e Ultrasound with 12L-RS wide band linear array 5-13 MHz frequency probe) under supervision by a radiologist. The procedure will be performed at the recovery bay in the operation theatre. The following measurements will be done and recorded for future statistical analysis.

• Depth of preepiglottic space (PE)
• Distances between epiglottis and midpoint between 2 vocal cords (E-VC)

1. Cormack Lehane score will be obtained during induction of General Anaesthesia and Intubation in OT on the specified OT date. Conventional direct laryngoscope will be used for CL score determination. Medical Officer of more than 2 years' experiences in anaesthesiology will be requested to intubate the patient to minimize bias. CL scores are as described below

CL 1: Most of glottis Visible CL 2: Half of the glottis seen CL 3: Only epiglottis visible CL4: No laryngeal structure visible

Study area: Hospital Sultanah Nurzahirah, Kuala Terengganu Study population: Adult patient age between 18 to 65, diagnosed or suspected OSA, planned for operation under General Anaesthesia

Sample size estimation

1. st Objective :

• To determine the correlation between airway ultrasonography measurements and CL score.
• Using R package pwr
• For PE
• r: 0.595 (based on Gupta D, 2012), significant level 0.05, power: 0.8
• Number of sample: 19
• For E-VC
• r: -0.966 (based on Gupta D,2012), significant level 0.05, power: 0.8)
• Number of sample: 4
• For ratio PE/E-VC
• r: 0.495 (based on Gupta D, 2012), significant level 0.05, power: 0.8)
• Number of sample: 28

1. nd and 3rdobjective

• 2nd objective: To determine the best cut off point of airway ultrasonography measurement in predicting the difficulty in intubation using CL (difficult vs less difficult)
• 3rd objective: To determine the accuracy of cut-off point in determining difficult intubation. (AUC) )
• Receiver Operating characteristics (ROC) curve power calculation using R
• Significance level: 0.05
• Power of study: 0.8
• Kappa: 4
• AUC: 0.7

Based on previous study, the incidence of difficult intubation in OSA patient is 19.3%. Hence kappa: 4. Using pROC package in R, based on the above values and data, the investigators need at least 24 cases of difficult intubation and 41 cases of non-difficult intubation (to detect AUC of 0.7 from 0.5 null hypothesis). So, total patients required is at least 65

Sampling method and subject recruitment As mentioned above, this is a cross sectional study. Involving face-to-face interview, USG airway assessment and Cormack Lehane score assessment. After the samples identified at the specific locations mentioned and they will be approached for consent. Once agreed, the patient will be recruited for the study.

Data analysis

Statistical Analysis Plan For objective 1

• Intended statistical analysis: Spearman correlation

For objective 2 and 3

• Using ROC curve
• outcome: difficulty status (0 or 1)
• test variable: PE, E-VC and ratio PE/E-VC

AUC in the ROC curve will be calculated accordingly based on the study results