The cricoid cartilage, as the narrowest part of the larynx in children, plays an important role in the selection of optimal ETT size for intubation [11,12]. However, recently, Dalal et al.  found that the vocal cord and subvocal cord areas were the narrowest portion in pediatric airways. Compared with the vocal cords, the cricoid is a complete and relatively rigid cartilaginous ring and the most frequently damaged structure during endotracheal intubation . Therefore, theoretically, the cricoid cartilage is the limiting factor during intubation and can be a predictive factor in the selection of optimal ETT size for intubation.
Ultrasonography is safe, noninvasive, can be quickly performed, provides real-time images, and is relatively simple to learn. More importantly, the leading edge of the cricoid cartilage and the air column can be identified with ultrasound. In our study, the cricoid cartilage diameters were estimated from the measurements of the transverse air column diameter, which is the most common approach to measure the transverse diameter of the cricoid cartilage in clinical research. To ensure the accuracy of the measurements, they were performed independently by two anesthesiologists and the average was taken, and the anesthesiologists who performed ultrasound examinations had performed 30 procedures before the formal experiment began. Lakhal et al. found that 15 procedures were enough for an operator to obtain reliable measurements. Moreover, Julio et al. found that subglottic diameter ultrasound measurements had high intra-rater and inter-rater reproducibility. Therefore, we believe that our measurements were accurate.
The leak test is a classic experimental method and has been applied to determine the best‑fit ETT size for many years. Therefore, in our study, best‑fit ETT size was chosen according to the leak test. In prior studies, the allowed leak pressure often was 15-30 cm H2O for cuffed ETT . Scoliosis can affect pulmonary function, and lung function abnormalities are mainly of the restrictive type . During the procedure, the children are positioned in prone recumbency, and the operation can apply pressure to the chest. All these factors can cause elevation of the airway pressure. Therefore, we chose a higher leak pressure of 25 cm H2O to determine the best‑fit ETT size.
Our study showed a strong correlation between the ETT size predicted by ultrasonography and the best‑fit ETT size in pediatric patients with thoracic or lumbar scoliosis. The Bland-Altman analysis showed no obvious bias between the ETT size predicted by ultrasonography and the best‑fit ETT size in pediatric patients with thoracic or lumbar scoliosis. Our findings were consistent with those reported by Pillai et al. . In their study, the correlation was 0.98, and the bias was 0.041 mm. Therefore, it is feasible to predict ETT size by measuring the transverse diameter of the cricoid cartilage with ultrasonography in pediatric patients with thoracic or lumbar scoliosis.
However, in pediatric patients with cervical lateral bending, there was a poor correlation between the ETT size predicted by ultrasonography and the best‑fit ETT size. Bland-Altman analysis showed that the average of the differences between the ETT sizes was 0.73 mm. Considering that the minimal increment for tube size change is 0.5 mm (according to the inner diameter of the ETT), the ETT size was overestimated by ultrasound in pediatric patients with cervical lateral bending.
MRI is considered the reference standard for evaluating the larynx. High-quality images of the cricoid cartilage can be acquired by MRI, and the cricoid cartilage diameter can be accurately measured . Therefore, in order to ascertain the cause of our result for Group C, we reviewed MRI images of four patients with cervical lateral bending. We found that the cricoid cartilage of the patients with cervical lateral bending was rotated (Figure 5). Previous research has shown that rotation of the centrum can produce displacement or rotation of the mainstem bronchi [9,10]. Therefore, we speculate that the rotation of the cricoid cartilage results from deviation or rotation of the cervical vertebrae.
Under normal circumstances, the cricoid cartilage is elliptical, and the transverse diameter is smaller than the anteroposterior diameter . When measuring the transverse diameter of the cricoid cartilage by ultrasound, the probe is positioned on the anterior neck, and the transverse air column diameter is measured to estimate the cricoid cartilage diameter. Because of this, rotation of the cricoid cartilage can broaden the air column as measured by ultrasound, resulting in a larger diameter measurement and overestimation of the necessary ETT size (Figure 6). In our study, we found that the ETT size predicted by ultrasonography was larger than the best‑fit ETT size in pediatric patients with cervical lateral bending. Therefore, compared to what is predicted by ultrasonography, these patients need a smaller ETT.
The incidence of laryngospasm was higher in the patients with cervical lateral bending. There are two possible reasons for this result. First, in pediatric patients with cervical lateral bending, the initial ETT size predicted by ultrasound was larger, and a larger ETT can irritate the throat and increase the likelihood of laryngospasm. Second, operation on the neck can irritate the upper airway, also increasing the likelihood of laryngospasm.
The present study has one limitation. Only four patients with cervical lateral bending underwent cervical MRI; therefore, it is not entirely clear whether the deviation of ultrasonographic measurements resulted only from the rotation of the cricoid cartilage in these patients. Further study is needed to analyze the influence of cricoid cartilage morphology on the accuracy of ultrasound to measure the cricoid cartilage diameter in pediatric patients with cervical lateral bending.