4 Discussion
TFBA is a clinical emergency and a major cause of death in children13. At present, the most effective way to remove TFBA is through the rigid bronchoscope as soon as possible. The most important diagnosis is based on the history of foreign body choking14.
LRTI is easily caused by TFBA. In this study, most children had purulent secretions. Previous studies reported that TFBA can cause obstructive emphysema and pneumonia15 based on preoperative CT. However, as a LRTI, acute bronchitis with no positive signs in CT is often ignored. Limited studies reported LRTI caused by TFBA, while they are mainly focused on bacterial culture7,8.Only one study reported the risk factors for LRTI10. As a result, the situation of LRTI cannot be fully showed.
In clinical practice, LRTI plays an important role in intraoperative operation and disease prognosis. A study reported that TFBA with LRTI have a longer hospital stay10. Therefore, we hope to find a relevant indicator to predict the occurrence of LRTI for operators so that they can pay more attention to prepare the operation. This study collected the children who were applied TFBA extraction and analyzed the risk factors for LRTI.
We used to have a clinical understanding that organisms are more likely to cause inflammation and suppuration in tracheobronchial tissues than non-organisms, especially plants16. In addition, it is reported that the type of TFBA was an independent risk factor for LRTI10. However, recent animal experiments found that there was no difference in tracheal damage between organic and inorganic foreign bodies9. Similarly, we found that there was no statistical difference in the incidence of LRTI among legumes, nuts, other organisms and non-organisms. We also found a correlation between the retention time of TFBA and LRTI. In our study the incidence of LRTI was 46.6%, 68.6% and 68.6% respectively with the retention time of within 24 hours, one week and more than one week,which was consistent with the results of a previous animal experiment9. In this animal experiment, it was found that most of the TFBA could cause tracheal damage within 5 days, and pneumonia more than 5 days. No matter in animal experiments or clinical findings, LRTI caused by TFBA could be aggravated with the increase of time.
We speculate that there are three main reasons for the differences between previous clinical understandings and our study. Firstly, the family members have a significantly higher understanding of TFBA than before, and the visiting time is significantly shortened. Secondly, CT has been more popular which increases the accuracy and timeliness of diagnosis. Thirdly, it may need time to realize the impact of the type of TFBA on trachea tissue. This suggests that no matter what type of TFBA is or where the TFBA is embedded, the longer the time is, the more serious the LRTI is. Therefore, we suggest that for children with long retention time of TFBA, the operators should be vigilant and fully prepared before operation. They should also inform the parents that the operation can be more difficult and complicated than those with short retention time.
In a report of LRTI caused by TFBA10, type, shape, the retention time of TFBA and age were independent risk factors for LRTI. Analyzed all the related risk factors in our study, except for the shape of TFBA, there was a correlation between the retention time of TFBA and LRTI, while there is no correlation between age, the type of TFBA and LRTI. We speculate that the main reason is that the family members have different understanding of TFBA, which leads to the difference in visiting time. In the above study, the median retention time of TFBA is 15 days, while it’s only 1 day in our study. It is controversial whether there is a difference in the degree of trachea inflammation caused by the type of TFBA. Although it has been reported that the intratracheal granulation in TFBA is related to the degree of TFBA oil release17, no correlation was found in animal experiments9. As for age, we speculate that the immunity of younger children is relatively weak so that they are easier to get LRTI when encounter the same TFBA. However, it needs time to show this difference.
No indicator of LRTI caused by TFBA was found in clinical symptoms. When choked by foreign body, the more severe the cough and asthma is, which indicates that the TFBA is too large, the larynx and trachea spasm caused by it is obvious and the blockage of foreign body is serious, the more dangerous the process of choking is. However, tracheal mucosal inflammation needs time to happen, these clinical symptoms cannot predict the incidence of LRTI. Despite of this fact, they reflect the severity of the disease. Operations should be done as soon as possible for children with obvious symptoms.
There is no specific standard for the subgroup classification of TFBA complications. According to previous literature, we found that there is similar classification for obstructive pneumonia and atelectasis in non-small cell lung cancer, which can be used as a reference. Firstly, non-small cell lung cancer respectively showed segmental and lobar local emphysema or atelectasis in incomplete or complete blockage of the bronchial lumen. Secondly, in the TNM stage of non-small cell lung cancer18,non-small cell lung cancer with obstructive pneumonia or partial or whole atelectasis is classified as T2 stage. In this guideline, atelectasis and emphysema are regarded as the same element. Furthermore, in previous prognostic studies of non-small cell lung cancer, preoperative obstructive pneumonia and atelectasis were also put into one group without discussed separately19when the overall survival rate and recurrence-free survival rate of lung cancer were compared. To sum up, we drew lessons from the classification method of non-small cell lung cancer, and put atelectasis and emphysema into the same group, considering them as different manifestations of the same type. Based on this method, we further divided atelectasis or emphysema and pneumonia into different subgroups. Combined with the clinical criticality, we classified the mediastinal emphysema, pneumothorax, subcutaneous emphysema, severe pulmonary inflammation, such as pleural effusion, and mediastinal displacement as a group. In univariate analysis, we found that the worse the clinical condition was indicated by cervicothoracic CT, the higher the incidence of LRTI would be. However, in multivariate analysis, there was an overlapping relationship between cervicothoracic CT diagnosis and the retention time of TFBA, suggesting that the longer the retention time was, the more serious the LRTI would be. The cervicothoracic CT diagnosis was not related to LRTI.
This study is limited in the following ways. Firstly, the number of cases included is small, and it is a single center study, which needs to be supplemented by a larger sample size. Secondly, this study only includes a single race. It is not clear whether there is a correlation between race and LRTI. Thirdly, the educational background of the families and their attention paid to children when they are eating may be risk factors for LRTI. Due to the lack of these information, this study is limited in its retrospective way.