Pre-Operative Airway Assessment of Patients With Morbid Obesity

Overview

Backround:The incidence of difficulty in laryngoscopy and endotracheal intubation is more common in obese patients. There are conflicting data on factors predicting difficult intubation. Also, these predictors are uncertain when video laryngoscopy (VL) is used. In this study we aim to investigate the correlation between airway assessment tests, anthropometric measurements and Modified Cormack-Lehane Classification (MCLC), assessed by C-MAC-D-blade VL and predictive values for intubation difficulty in class-III patients. Methods: This prospective observational study was performed in 121 obese patients (Body mass index (BMI)≥45kg/m2) undergoing bariatric surgery. BMI, modified mallampati scores (MMS), thyromental distance (TMD), sternomental distance (SMD), interincisor distance (IID), and neck, waist, and chest circumference were all recorded preoperatively. Endotracheal intubations were performed with a C-MAC D-blade VL. The glottic view and intubation difficulty were determined according to the MCLC. MCLC I-IIa was recorded as the group with no risk of a difficult intubation, the MCLC IIb-III was recorded as the group with risk of a difficult intubation and the MCLC-IV was recorded as the group with a difficult intubation. The correlation between the airway tests, anthropometric parameters and the MCLC, as well as their sensitivity and specificity in predicting the intubation difficulty based on the MCLC, were analyzed. The duration of intubation, number of attempts, external laryngeal maneuver, the equipment requirements (style, guide), and complications during intubation were recorded.

Study Design This prospective observational clinical study was conducted in patients with a BMI ≥ 45 kg/m2 undergoing elective bariatric surgery between January 2018 and January 2019, in the Anesthesiology and Reanimation Clinic of the Health Sciences University, Bagcilar Training and Research Center, after obtaining approval from the Ethics Board of Health Sciences University Istanbul Bagcilar Training and Research Hospital (17.08.2017-2017/599). Patient population The number of participants in this study was calculated based upon a previous study. A sample size calculator showed that 116 participants would be required to predict difficult intubation in obese patients (with 80% power, α=0,05 and, β=0,2). After obtaining the patients written informed consent, as a result of the power analysis, one hundred and twenty-one patients, aged 16-68 years, with a BMI ≥ 45 kg/m2, who were undergoing elective bariatric surgery were included in this study. Patients with a history of difficult airway, limited neck movements, oral-pharyngeal cancer or reconstructive surgery, cervical spinal injury and facial anomaly or scar, patients who required quick and awake endotracheal intubation, and patients who were not oriented-cooperative, were excluded from the study. Study Protocol Age, sex, height, body weight, BMI, airway assessments and anthropometric measurements, including the MMS, thyromental distance TMD), sternomental distance (SMD), interincisor distance (IID), and neck, waist and chest circumference measurements of all patients were evaluated and recorded preoperatively. The MMS was assessed by Samsoon and Young Modifications and scores were assessed 1 to 4. The TMD and SMD were measured with the neck fully extended. The TMD was measured between the upper border of the thyroid cartilage and the bony point of the mentum, and the SMD was measured between the upper border of the manubrium sterni and the bony point of the mentum. The IID was measured when the patient opened their mouth, and the distance between upper and lower incisors was obtained. Neck circumference was measured at the level of the thyroid cartilage when the head was in a neutral position. The chest and waist circumference was measured around the widest circumference. After patients were admitted to the operating room, all patients were positioned with the gel placed under their head. The intravenous route was inserted in the dorsum of the hand using a 20 G (Gauge) venous cannula, and a 2-4 mL/kg balanced crystalloid was infused. All patients were premedicated with 1-2mg midazolam (Zolamid 15mg/3ml, Defarma, Turkey). Each patient was routinely monitored with the standard monitorization, including three-lead electrocardiography (ECG), non-invasive arterial pressure, peripheral oxygen saturation (SpO2), and neuromuscular monitorization with the TOF-Watch SX (Organon, Ireland) device. All patients were pre-oxygenated using a face mask, with 100% oxygen for 3 to 5 minutes prior to induction of the anesthesia. Induction is made with 2 mg/kg propofol (Propofol 1%, Fresenius kabi, Germany) and 1 µgr/kg fentanyl (Talinat, VEM, Turkey) based on a lean body weight, and 0.6 mg /kg rocuronium bromide (Esmeron Organon, Holland) based on an ideal body weight. Following the adequate paralysis (loss of responsiveness to train of four stimulations with TOF), all patients were routinely intubated using C-MAC D-blade VL (Storz 8402 ZX C- MAC Karl Storz, Germany). Anesthesia was maintained with inhalation of 50% oxygen-medical air mixture and 2% minimum alveolar concentration (MAC) of sevoflurane (Sevoflurane, Abbott, England). The time of the tracheal intubation was defined from insertion of the VL in the mouth until the endotracheal tube was placed through the vocal cords and expired carbon dioxide at capnography was observed. In the case of a peripheral oxygen saturation decrease (\< 90%), performing of intubation was interrupted, and mask ventilation was resumed. As there is no validated classification for the glottic view with VL intubation, it was evaluated and recorded according to the MCLC modified by Yentis and Lee, without any laryngeal pressure as shown below. MCLC I: A full view of the glottis MCLC IIa: Partial view of the glottis is visible MCLC IIb: Arytenoids or posterior part of the vocal cords are only just visible MCLC III: Only the epiglottis is visible MCLC IV: Neither epiglottis nor the glottis are visible MCLC I-IIa was recorded as the group with no risk of a difficult intubation, MCLC IIb-III was recorded as the group with a risk of a difficult intubation, the grade MCLC IIb was associated with a greater difficulty for intubation with a grade III, and MCLC-IV was recorded as the group with difficult intubation. The time of intubation, number of attempts, external laryngeal maneuver, stylet or both the usage requirements and complications (airway trauma, soft tissue hemorrhage, tooth damage, etc.) were recorded. All airway assessment tests, and anthropometric parameters were performed by the same anesthesiologist. All the endotracheal intubations and assessment of MCLC were routinely performed with a C-MAC D-blade VL by another anesthesiologist who was blinded to the datas of the preoperatively assessed tests. Primary and Secondary Outcomes The primary outcomes were the correlation between the preoperative airway assessment tests (MMS, TMD, SMD, IID), anthropometric measurements (BMI, neck, waist, chest circumference), and MCLC assessed by VL as well as their predictive values for the intubation difficulty according to the MCLC. Secondary outcomes were the time for intubation, number of attempts, equipment (stylet) and external laryngeal maneuver (ELM) requirements, as well as the complications during intubation (lip trauma, soft tissue hemorrhage, tooth damage, supraglottic trauma, etc.). Statistical Methods Statistical analyses were performed using SPSS Statistics 21 Statistical Package for the Social Sciences (SPSS Inc, Chicago, IL, USA). Continuous variables are expressed as means and standard deviation. The Kolmogrov-Smirnov test was used to investigate the normal distribution of continuous data. The Student's test was used for normally distributed continuous variables. The Mann-Whitney U test was used for variables that were not normally distributed. The Chi-Square and Fisher's Exact tests were used for non-continuous variables. To evaluate the change of continuous variables both Pearson correlation analysis was used for parametric data and Spearman Correlation analysis was used for nonparametric data. The evaluation of correlation coefficient; r=0.00-0.24, r=0.25-0.49, r=0.50-0.74, r=0.75-1.00 were considered as weak, moderate, strong and very strong respectively. The cut-off points for statistically significant parameters were determined by using the receiver operating curve (ROC) analysis which resulted in the best combination for sensitivity and specificity. Sensitivity, specificity, positive predictive and negative predictive values were calculated in the presence of significant limit values. All variables at p \< 0.05 were considered as statistically significant.