The main findings of this study were that (1) single-lumen tracheal tube fixation, where the vocal cord guide overlaps with the patient’s vocal cord, has a risk of incomplete sealing of tracheal diverticula depending on the tracheal tube products and (2) no products in this study satisfied both outcomes of a low risk of incomplete sealing of tracheal diverticula (< 5%) and a low risk of unintended bronchial intubation (< 5%).

Tracheal diverticula were previously considered a rare congenital or acquired condition1; however, but because of advances in diagnostic imaging, such as thin-slice CT, recent studies have reported the prevalence of tracheal diverticula to be 0.3%–8.1% and that it is not rare [1, 3, 17,18,19,20,21]. In our study, the prevalence of tracheal diverticula was 5.7%, which is similar to that reported previously [1, 3, 17,18,19,20,21]. According to these reports, it is thought that many tracheal intubation attempts were carried out in patients with tracheal diverticula without the practitioner noticing the existence of tracheal diverticula. In this study, tracheal diverticula were commonly located on the right dorsal side of the trachea at the sternoclavicular joint level, which almost coincided with the level of the second thoracic vertebral body; this finding is consistent with previous findings [1, 2]. The trachea at that level is the transitional point between the extrathoracic and intrathoracic trachea. At this level, the right dorsal wall of the trachea lacks the support of the esophagus and is considered one of the most vulnerable regions in the trachea against positive airway pressure anatomically [2].

In general anesthesia, positive-pressure ventilation for tracheal diverticula could causes tracheal diverticula rupture, subcutaneous emphysema, and pneumomediastinum [7,8,9,10,11,12,13,14]. Thus, tracheal diverticulum is considered as a comorbidity that requires attention in airway management with positive pressure ventilation and preventing positive-pressure ventilation to tracheal diverticula is important to avoid these adverse events. Many of the patients with tracheal diverticula in this study did not have an evident orifice, and the possibility that some patients did not have a tracheal diverticulum but had a paratracheal air cyst could not be ruled out. However, a dynamic CT report has shown that even in paratracheal air cysts without an evident orifice, the cyst contracts during inspiration and expands during expiration, indicating that the many paratracheal air cysts without an evident orifice might be tracheal diverticula and are actually in communication with the trachea [2]. Therefore, the absence of an evident orifice on CT does not mean that there is no diverticulum-tracheal communication. The intra-diverticular pressure during ventilation depends on the diameter of orifice and the existence of a diverticulum-tracheal communication. However, many of the patients in this study did not have large orifice diameters, which may be related to relatively low incidence of injuries related to tracheal intubation and positive pressure ventilation compared to the high prevalence of tracheal diverticulum.

Most tracheal tubes assessed in this study had a risk of incomplete sealing of tracheal diverticula during tracheal intubation with fixation of the tracheal tube where the vocal cord guide overlaps with the patient’s vocal cord. High positive pressure ventilation could widen the diverticulum-tracheal communication that is closed or very small during spontaneous ventilation and could dilate tracheal diverticula, and we therefore consider that all orifices should be sealed or bypassed by the tracheal tube during positive pressure ventilation to avoid positive pressure ventilation into the tracheal diverticula. However, the possibility that paratracheal air cyst without evident orifice has hidden orifices and the possibility that tracheal diverticulum with evident orifice has other hidden orifices cannot be ruled out [2]. Therefore, to prevent positive pressure ventilation to the tracheal diverticulum during positive pressure ventilation in a clinical setting, it is important to seal and bypass not only the obvious orifice but also the entire tracheal diverticulum by the tracheal tube because the positional relationships between all of the orifices, including hidden orifices, and the trachea are not known.

Preventing the incomplete sealing of tracheal diverticula involves complete sealing by the tracheal cuff by fixing the tube deeply where the tracheal cuff is located completely distal to the tracheal diverticula [15]. This method needs deeper tube fixation than it would be in accordance with the vocal cord guide and has a risk of unintended bronchial intubation [22,23,24]. Although the best way to avoid these risks is to measure the patient’s tracheal length and the characteristics of the tracheal diverticula, to select the most appropriate tracheal tube and inner diameter size, and to use a bronchoscope for confirming no bronchial intubation during general anesthesia in all cases, this requires a significant effort and is impractical. We presented the indications for low-risk tracheal tubes and tube fixation in reference to incomplete sealing of tracheal diverticula and unintended bronchial intubation. When the tube was fixed deeper than in accordance with the vocal cord guide, a length from the distal end of the tracheal cuff to the patient’s vocal cord of ≥ 70 mm had a low risk of incomplete sealing of tracheal diverticula. When the tube was fixed shallower than in accordance with the vocal cord guide, a length from the distal end of the tube to the patient’s vocal cord of ≤ 95 mm had a low risk of unintended bronchial intubation. These indications are beneficial for anesthesiologists when deciding the type and fixation depth of tracheal tubes. Although this was a single-center study performed in Japan and all patients were Asian, the mean tracheal length was similar to that in other reports from Japan and those from other countries, such as England, New Zealand, and East Asian countries [25,26,27,28,29,30], and the mean tracheal length might be similar across populations. Thus, the characteristics and positional relationships between a tracheal diverticulum and the tracheal tube in this study could be extrapolated to other populations.

Taken together, anesthesiologists might have to consider individual structural parameters of the trachea and tracheal diverticula and choose appropriately sized tracheal tubes to avoid the compound risk of incomplete sealing of tracheal diverticula and unintended bronchial intubation in patients with tracheal diverticula. The results and indications in our study may help anesthesiologists make these decisions.

This study has several limitations. First, this study did not assess other tracheal lesions. Bronchial diverticula are some of the most identified tracheal outpouching lesions whose prevalence were reported to be 21.6%–45.5%; most bronchial diverticula are located at the carina and main bronchi [31,32,33]. Although almost all bronchial diverticula are located at the distal end of the tracheal tube in the tracheal intubation and numerous bronchial diverticula have received positive-pressure ventilation to date, no study has reported the rupture or other issues of bronchial diverticula related to tracheal intubation. Thus, the impact of positive-pressure ventilation on bronchial diverticula might be lower than that on tracheal diverticula. Second, this study did not assess the classification of tracheal diverticulum as congenital or acquired and the changes in tracheal diverticula over time or the direct impact of positive-pressure ventilation to tracheal diverticula that were sealed incompletely. Further research is needed to investigate the natural course of tracheal diverticula and their changes after general anesthesia. Third, other types of tracheal tubes, which were not assessed in this study, may have different structural parameters and positional relationships between tracheal diverticula and the tracheal tube. Forth, this study is an observational and simulation study using thoracic CT. In the general anesthesia, tracheal length may change according to the patient’s position (e.g., lateral position or head-down tilt) and patient’s neck position (anteflexion or retroflexion). Despite these limitations, to the best of our knowledge, this study is the first to investigate the positional relationships between tracheal diverticula and the tracheal tube with tracheal intubation during general anesthesia. Anesthesiologists perform tracheal intubation routinely. Thus, the present findings raise the importance of caution regarding incomplete sealing of tracheal diverticula and the risk of positive-pressure ventilation to tracheal diverticula.