The Journal of Laryngology and Otology October 1990, Vol. 104, pp. 778-782 Radiological diagnosis of aspirated foreign bodies in children: Review of 343 cases LIANCAI Mu, M.D., DEQIANG SUN, M.D., PING H E , M.D. (Shenyang, China) Abstract In our series of 400 Chinese children with foreign body aspiration (FBA), 343 cases were evaluated by fluoroscopy and/or plain chest X-rays before endoscopic removal of the foreign bodies. The majority of the foreign bodies (FBs) were organic (378/400, 94.5 per cent). The results showed that mainstem bronchial foreign bodies were diagnosed correctly in 68 per cent of cases compared with 65 per cent correct diagnoses with segmental bronchial foreign bodies, but only 22 per cent correct diagnoses with tracheal, and 0 per cent correct diagnosis in those with laryngeal foreign bodies. Eighty per cent (32/40) of the children with laryngotracheal FBs had normal X-ray findings, whereas 67.7 per cent (205/303) of the children with bronchial FBs had abnormal chest X-ray findings. The most common positive radiological signs in the children with tracheobronchial FBs were obstructive emphysema (131/213,62 per cent) and mediastinal shift (117/ 213,55 per cent). The incidence of major complications was related not only to the size of the foreign body and its location but also the duration since aspiration. The most common types of bronchial obstructions by airway FBs are discussed. per cent) (Kim et al., 1973; Aytac et al., 1977; Li et al., 1980; Rothmann and Boeckman, 1980; Banerjee et al., 1988; McGuirt et al., 1988), pneumonia (8-20 per cent) (Kim etal., 1973; Li etal., 1980; Rothmann and Boeckman, 1980; McGuirt et al., 1988), normal findings (6.1-24 per cent) (Kim et al., 1973; Aytac et al., 1977; Rothmann and Boeckman, 1980; Wiseman, 1984; Banerjee et al., 1988; McGuirt et al., 1988), and radiopaque FB (6-17 per cent) (Kim et al., 1973; Aytac et al., 1977; Rothmann and Boeckman, 1980; McGuirt etal., 1988). The incidence of complications is also related to the duration of FBA. Esclamado and Richardson (1987) reported that the incidence of major complications was 45 per cent; however, in patients with a delay in diagnosis of over 24 hours the complication rate was 67 per cent. Therefore, early diagnosis and treatment can decrease the number and severity of complications. The clinical presentation, diagnosis, and treatment of FBA in children have been well reviewed and documented in recent literature. However, a detailed retrospective analysis of the radiological evaluation of FBA is still needed. This study highlights the radiological, especially fluoroscopic, features of airway obstruction due to FB. We also attempt to detect the relationship between the FB's size and X-ray findings, and to describe some aspects of the most common types of bronchial obstructions by airway FBs. Introduction Foreign body aspiration (FBA) is a very serious and lifeendangering emergency which is quite frequently encountered in small children. Diagnosis of airway FBs in children is still an important problem. It is well known that a diagnosis of FBA can be made based on a positive history of aspiration and particular clinical manifestations such as cough, wheeze, respiratory distress, and decreased air entry. However, the history of aspiration is sometimes not typical. It has been reported that a positive history of FBA cannot be obtained 3-27 per cent of the time (Kim et al., 1973; Li et al., 1980; Rothmann and Boeckman, 1980; Wiseman, 1984; Esclamado and Richardson, 1987; Banerjee et al., 1988). In nearly one third of children aspirating FBs, the actual event is not witnessed (Cohen et al., 1980). In addition, physical examination may show no abnormality in some cases. Therefore, radiological evaluation has widely been used in the diagnosis of FBA clinically although a normal chest X-ray film was obtained by Aytac et al. (1977) in about one quarter of such cases. Many investigations have demonstrated that the majority of the FBs (66.4-87 per cent) retrieved at endoscopy are organic matter (Kim et al., 1973;Liera/., 1980; Banerjee et al., 1988). A radiopaque FB is readily identified, but a radiolucent one can be suspected only from the secondary effects it produces on the lungs. The commonest radiographic findings in the published reports have been mediastinal shift (4—71 per cent) (Aytac et al., 1977; Li et al., 1980), obstructive emphysema (17-60 per cent) and atelectasis (12-34.4 Subjects and methods During the seven years from June 1982 to Novembet From the Department of Otolaryngology, 3rd Affiliated Hospital, China Medical University, Shenyang, Liaoning, People's Republic of China. Accepted for publication: 31 July 1990. 778 779 RADIOLOGICAL DIAGNOSIS OF ASPIRATED FOREIGN BODIES IN CHILDREN TABLE I TABLE II LODGEMENT SITE OF ASPIRATED FOREIGN BODIES X-RAY FINDINGS IN FOREIGN BODY ASPIRATION Cases Cases Site No. % X-ray No. % Larynx Trachea Right mainstem bronchus Left mainstem bronchus Right segmental bronchi Left segmental bronchi 4 48 150 117 36 45 1 12 38 29 9 11 Positive Negative No X-ray 213 130 57 53 33 14 Total 400 100 Total 400 100 1989 inclusive, 400 children with FBA were treated at the Department of Otolaryngology, 3rd Affiliated Hospital, China Medical University, Shenyang, People's Republic of China. The medical charts of the 400 cases of FBs in the airway were reviewed. We noted the age and sex of each; the nature, size, and location of the FBs; the time lag between the aspiration and diagnosis; the results of radiological examination on admission; and endoscopic findings. Radiological investigation consisted of chest fluoroscopy in 264 (66 per cent) children, plain chest X-rays in 25 (6.3 percent) cases, and a combination of fluoroscopy with plain chest X-ray film in 54 (13.3%) cases. Fiftyseven children (14.4 per cent) did not require X-ray examination before endoscopy because they had a definite history of aspiration and had to be operated upon as emergencies. Results Of the 400 children, 359 (89.8 per cent) were under three years. Their ages ranged from 7 months to 13 years. The malerfemale ratio was 1.2:1. Table I shows the various locations of the FBs in our 400 cases. Laryngotracheal FBs were relatively uncommon. Of the bronchial FBs, right-side FBs were more common (186/400, 47%) compared with left-side FBs (162/400,40 per cent). The FBs were most often found in the mainstem bronchi (267/400, 67 per cent). The majority of the FBs were organic (378/400, 94.5 per cent), 61 per cent of them being peanuts (232/378). Bean and sunflower seeds were next most common, seen in 101 cases (26.7 per cent, 101/378). The remainder included corn (20/400, 5 per cent), watermelon seeds (17/400,4.2 per cent), pieces of fruits and vegetables (8/ 400, 2 per cent). Among the inorganic substances, plastic and metallic objects, glass, eggshell, piece of bone, etc., were encountered (22/400, 5.5 per cent). The radiological findings encountered are summarized in Tables II, III, IV, and V. Over one third (130/343, 38 per cent) of the children with airway FBs had normal X-ray findings (Table II). Eighty per cent (32/40) of the children with laryngotracheal FBs had normal X-ray findings, whereas 67.7 per cent (205/303) of the children with bronchial FBs had positive X-ray findings (Table III). Tables IV and V show the positive X-ray findings in 213 cases with airway FBs. As Table V shows, the most common complication of FBA was obstructive emphysema (131/213, 62 per cent). Inspiratory shift of the mediastinal shadow was seen in 55 per cent of the cases, pneumonia in 26 per cent, and atelectasis in 18 per cent. More than one of these positive findings were present in some cases. Radiopaque FBs were seen in 7 cases (3 per cent). In those children with mainstem bronchial FBs, 40 per cent (106/267) had obstructive emphysema, and 35 per cent (94/267) had inspiratory shift of mediastinal shadow (Tables I, V). Table VI shows the mean duration of aspiration of FBs in 400 cases. The mean duration of aspiration of bronchial FBs was the longest. Table VII shows the relationship between X-ray findings and interval since aspiration of FBA. Considerable differences in X-ray findings were observed between children diagnosed early and late. In patients with early diagnosis of less than 24 hours the incidence of complications was 44 per cent (28/63). In contrast, the complication rate in those patients who had a delay in diagnosis of more than 24 hours was over 60 per cent. In 21 cases the time interval from either known aspiration or onset of symptoms to diagnosis was more than 30 days. All but one (20/21, 95%) had positive radiological findings. The complication rate increased with the length of FB's sojourn in the airways. TABLE III CORRELATION OF X-RAY FINDINGS WITH ENDOSCOPY AND REMOVAL OF FOREIGN BODIES FROM THE AIRWAYS Foreign body location Trachea Larynx X-ray No. (%) Mainstem bronchi No. (%) No. (%) Segmental bronchi No. (%) Positive Negative No X-ray 0 3 1 (0) (75) (25) 8 29 11 (17) (60) (23) 160 74 33 (60) (28) (12) 45 24 12 (55) (30) (15) Total 4 (100) 48 (100) 267 (100) 81 (100) 780 LIANCAI MU, DEOIANG SUN AND PING HE TABLE IV DISTRIBUTION OF POSITIVE X-RAY FINDINGS IN 213 CASES Cases Location No. % Trachea Mainstem bronchi Segmental bronchi 8 160 45 4 75 21 Total 213 100 Table VIII shows the correlation of negative X-ray findings with duration of FBA. More than one half (56 per cent) of the patients with early diagnosis (=S24 hours) had normal X-rays. By comparison, only one third of patients with a delay in diagnosis of over 24 hours had normal chest X-rays. Table IX shows the location, type and size of organic airway FBs with negative X-ray findings. Small pieces of peanut (67/130, 52 per cent), sunflower seed and. watermelon seed (25/131, 19 per cent), and small piece of bone (18/130, 14 per cent) were the most frequently found airway FBs with negative X-ray findings. Discussion Although ultrasonography, CT scanning, and xeroradiographic examinations have all been advocated in the diagnosis of aspirated FBs, they are of limited usefulness. Clinically, chest fluoroscopy and X-rays that include inspiratory and expiratory, anteroposterior and lateral films, have been used widely by many investigators. We prefer to use fluoroscopy as a routine diagnostic procedure. In our series, 80 per cent of the cases were examined fluoroscopically before bronchoscopy. We have found that fluoroscopy plays an important role in the diagnosis of organic bronchial FBs. As our results show, fluoroscopy increases the accurate diagnosis in over two-thirds of the cases of bronchial FBs, whereas the discrepancy between the fluoroscopic and bronchoscopic findings is less than one third. We feel that fluoroscopy is much better than chest X-rays in determining whether there is abnormal behaviour of the mediastinal shadow. In some cases with a normal chest X-ray, inspiratory-expiratory mediastinal shift indicating a unilateral bronchial obstruction was observed by means of fluoroscopy. Although fluoroscopy is the classic laboratory diagnostic measure used, it has n o r been uniformly helpful. In this study, 80 per cent of the children with laryngotracheal FBs had normal X-ray findings. Therefore, one must pay attention to the history of aspiration and clinical manifestations in those cases with normal X-ray findings. Meanwhile, we have to accept the fact that in determining whether mild pneumonia or atelectasis caused by aspirated FBs is present chest X-rays seem better than fluoroscopy. In some cases with normal fluoroscopic findings, complications such as pneumonia and/or atelectasis were discovered only on X-ray. Recurrent pneumonia as a common complication is usually caused by a long-standing bronchial FB. The combination of fluoroscopy with X-ray is very useful for those cases with suspicious or unusual fluoroscopic findings. As our results show, main bronchial FBs are diagnosed correctly 68 per cent of the time compared with 65 per cent correct diagnoses with segmental bronchial FBs and with only 20 per cent correct diagnoses with laryngotracheal FBs. Our findings are similar to those (86.4 per cent, 66.7 per cent, and 35 per cent, respectively) presented by Jiang et al. (1984). We found that the X-ray findings are related to the following factors: the size of the FB (Table IX), the site of its lodgement (Tables III, IV, V), and the length of its sojourn in the airway (Tables VII, VIII). The development of the pulmonary complications by FBA is related to the pattern of bronchial obstruction, whereas that of airway obstruction depends upon the FB's size relative to that of the bronchus in which it lodges. Four types of bronchial obstructions by airway FBs, namely, the check-valve, stop-valve, ball-valve, and bypass-valve obstructions, have been described by Chatterji and Chatterji (1972). According to the description of airway obstructions, the commonest radiological findings produced by aspirated FBs may be explained as follows: 1. Normal X-ray findings The bypass-valve obstruction, caused by the organic FBs which partially obstruct the respiratory tract (larynx, trachea, or bronchus) in both phases of respiration, leads to diminished aeration and opacity on the affected side. If there is no marked pressure difference on the two sides of the mediastinum, mediastinal shift cannot be observed. Therefore, no abnormality appears on both the X-rays and fluoroscopy. Organic FBs with negative TABLE V POSITIVE X-RAY FINDINGS IN 2 1 3 CASES WITH FOREIGN BODY ASPIRATION Site of foreign bodies Main bronchi (N=160) Trachea (N=8) X-ray findings Obstructive emphysema Mediastinal shift Pneumonia Atelectasis Radiopaque object Segmental bronchi (N=45) Total No. (%) No. (%) No. (%) No. (%) 4 4 2 0 0 (2) (2) (1) (0) (0) 106 94 35 24 5 (50) (44) (16) (11) (2) 21 19 19 15 2 (10) (9) (9) (7) (1) 131 117 56 39 7 (62) (55) (26) (18) (3) *More than one of these positive findings were present in some cases. 781 RADIOLOGICAL DIAGNOSIS OF ASPIRATED FOREIGN BODIES IN CHILDREN from it on expiration. In this case, no abnormality appears on the X-rays, but fluoroscopy can show mediastinal shift. TABLE VI MEAN DURATION OF FOREIGN BODY ASPIRATION IN 4 0 0 CASES Duration. Location of foreign bodies Mean days (range) Larynx (N=4) Trachea (N=48) Mainstem bronchi (N=267) Segmental bronchi (N=81) 3.8 (1-8) 4.6 (1 hr-17) 10.9 (30 min-365) 14.3 (2 hrs-120) X-ray findings are usually small in size and flat or oblong in shape, such as sunflower or watermelon seeds, eggshell, corn, and small pieces of bean or peanut (Table IX). 2. Obstructive emphysema The check-valve obstruction, by which air can be inhaled and cannot be expelled, gives rise to obstructive emphysema on the involved side. It has been wellknown that the diameter of a bronchus increases during inspiration and decreases during expiration. The FB may be of a size that it occludes the bronchus during expiration but not during inspiration; thus it acts like a valve, so that the lung beyond the obstruction becomes inflated. Poor airflow out of an obstructed lung or lobe will result in unequal emptying and will be visualized as obstructive emphysema on radiological examination, and the FB can be located in the supplying bronchus. 3. Mediastinal shift The types of bronchial obstructions which can cause mediastinal shift are as follows: (1) The ball-valve obstruction, which dislodge during expiration and reimpact during inspiration, leads to early atelectasis. In the latter cases, the mediastinal shift towards the affected side can be observed. This type of obstruction is named as 'ball-valve obstruction' because it is usually associated with rounded smooth FBs. (2) The check-valve obstruction can also cause early mediastinal shift towards the opposite side. (3) The bypass-valve obstruction may give rise to slight mediastinal shift. With the bronchus partially occluded, less air flows in and out. If the pressure difference on the two sides of the mediastinum is created, it causes the mediastinum to swing from side to side—towards the affected lung on inspiration and away 4. Atelectasis The types of bronchial obstructions which can cause atelectasis are as follows: (1) The stop-valve obstruction, associated with obstruction of a large air passage in both inspiration and expiration, results in collapse or consolidation of the affected bronchopulmonary segment. This may be caused either by a large FB which leads to complete occlusion from the time of aspiration or by a gradually swollen, small, organic body which had previously resulted in a check valve type of obstruction. In this case, all air in the involved part of the lung will be gradually absorbed, leaving the lung airless and collapsed. The mediastinal shift towards the affected side can be observed in the latter cases. The chest X-rays will show signs of atelectasis. The FB can then be found in the supplying bronchus. (2) The ball-valve obstruction can also cause early atelectasis. 5. Pneumonia Each type of the bronchial obstructions by aspirated FBs mentioned above can cause pneumonia if the impaired outflow of secretions has persisted for several days or weeks. 6. Pneumothorax This complication caused by aspirated FBs is uncommon. It may result either from the obstructive emphysema or from the penetration through a bronchial wall of a sharp FB. It has been accepted that airway FBs should be strongly suspected in children who present with a suggestive history, even when physical findings and radiological studies are negative. One must remember the fact that more than half of the children with laryngotracheal FBs have normal radiological findings and this may lead one to a false sense of security. The presentation and complications of FBs can be determined by the history of aspiration, physical examination, and radiological studies. Bronchoscopy of the airway is an TABLE VII RELATIONSHIP BETWEEN X-RAY FINDINGS AND DURATION OF FOREIGN BODY ASPIRATION IN 3 4 3 CASES X-ray findings 24hrs 1-3 days 4-7 days 8-14 days 15-30 days 31-365 days (N=63) (N=94) (N=75) (N=34) (N=56) (N=21) Negative (N= 130, 38%) Positive (N=213, 62%) obstructive emphysema mediastinal shift pneumonia atelectasis 35 38 27 12 17 1 19 20 3 2 35 34 10 4 26 25 11 11 18 12 8 5 23 18 16 9 10 8 8 8 Complication rate 44% 60% 64% 65% 70% 95% *More than one of these positive X-ray findings was present in some cases. 782 LIANCAI MU, DEQIANG SUN AND PING HE TABLE VIII CORRELATION OF NEGATIVE X-RAY FINDINGS WITH DURATION OF FOREIGN BODY ASPIRATION IN 1 3 0 CASES Cases with negative X-ray findings (days) X-ray examination No. % 63 94 75 34 56 21 35 38 27 12 17 1 56 40 36 35 30 5 <1 1- 3 4- 7 8-14 15-30 >30 3. The combination of fluoroscopy with chest X-rays is very useful for some cases with suspicious or unusual fluoroscopic findings. 4. Over two-thirds of the children with laryngotracheal FBs had normal X-ray findings in this series. Therefore, a negative X-ray finding cannot rule out the diagnosis of FBA. 5. The positive X-ray findings of an organic FB are related to six factors, namely, the size and shape of FB; the site of its lodgement; the pattern of bronchial obstruction; the length of its sojourn in the airways; and the technique or radiographic evaluation used. 6. The diagnosis of airway FB should not be dismissed without bronchoscopy. TABLE IX DISTRIBUTION OF 1 3 0 ORGANIC FOREIGN BODIES WITH NEGATIVE X-RAY FINDINGS IN THE AIRWAYS Acknowledgement Cases Location of foreign bodies Type and size of the foreign bodies No. Larynx i bean piece of eggshell Trachea i-i peanut 1 sunflower seed 1 watermelon seed 1 bean piece of eggshell 1 corn miscellaneous 11 4 4 3 2 2 3 Mainstem bronchi ^—i peanut ibean i-1 sunflower seed j-1 corn 1 watermelon seed miscellaneous 39 14 9 4 3 5 Segmental bronchi small piece of peanut i sunflower seed miscellaneous 1 2 3 29 74 Total (%) This research was supported by Postgraduate Research Grants from China Medical University, Shenyang, P.R. of China. (2) References Aytac, A., Yurdakul, Y., Ikizler, C , Olga, R., Saylam, A. (1977) Inhalation of foreign bodies in children: Report of 500 cases. Journal of Thoracic and Cardiovascular Surgery, 74: 145—151. (22) (57) 17 5 2 24 (19) 130 (100) effective way of establishing the diagnosis of FBA. In any suspected cases, bronchoscopic evaluation should be performed as early as possible to diminish severe complications by aspirated FBs. Conclusions The results obtained from the study are contributory to the following conclusions: 1. Radiological examination is essential for detection and localization of FBs in the airways. However, the physician must be aware of the advantages and pitfalls of radiological diagnosis of aspirated FBs. 2. Fluoroscopy plays an important role in the diagnosis of organic bronchial FBs. In determining whether there is abnormal behaviour of mediastinal shadow fluoroscopy is much better than chest X-rays. Banerjee, A., Subba Rao, K. S. V. K., Khanna, S. K., Narayanan, P. S., Gupta, B. K., Sekar, J. C , Retnam, C. R., Nachiappan, M. (1988) Laryngo-tracheo-bronchial foreign bodies in children. Journal of Laryngology and Otology, 102: 1029-1032. Chatterji, S., Chatterji, P. (1972) The management of foreign bodies in air passages. Anesthesia, 27: 390-395. Cohen, S. R., Herbert, W. I., Lewis, G. H., Geller, K. A. (1980) Foreign bodies in the airway: five-year retrospective study with special reference to management. Annals of Otology, Rhinology and Laryngology, 89: 437-442. Esclamado, R. M., Richardson, M. A. (1987) Laryngotracheal foreign bodies in children: A comparison with bronchial foreign bodies. American Journal of Diseases of Children, 141:259-262. Jiang, X. L., Liu, J. B., Yang, X. J. (1984) The radiological diagnosis of vegetable foreign body of trachea and bronchi in children. Ada Academiae Medicinae Jinzhou, 5: 11-15. Kim, I. G., Brummitt, W. M., Humphry, A., Siomra, S. W., Wallace, W. B. (1973) Foreign body in the airway: A review of 202 cases. Laryngoscope, 83: 347-354. Li, C. F., Han, J. C , Xu, G. J. (1980) Experience in the treatment of foreign bodies in the trachea and bronchus: With an analysis of 533 cases. Beijing Medical Journal, 2: 68-70. McGuirt, W. F., Holmes, K. D., Feehs, R., Browne, J. D. (1988) Tracheobronchial foreign bodies. Laryngoscope, 98: 615-618. Rothmann, B. F., Boeckman, C. R. (1980) Foreign bodies in the larynx and tracheo-bronchial tree in children: A review of 225 cases. Annals of Otology, Rhinology and Laryngology, 89: 434-436. Wiseman, N. E. (1984) The diagnosis of foreign body aspiration in childhood. Journal of Paediatric Surgery, 19: 531-535. Address for correspondence: Liancai Mu, M.D., Lecturer, Sahlgren's Hospital, ENT Clinic, University of Goteborg, S-413 45, Goteborg, Sweden.