- Email: [email protected]
A heuristic approach to foreign bodies in the paediatric airway
G Model
PEDOT-7336; No. of Pages 5 International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
A heuristic approach to foreign bodies in the paediatric airway Dora Blair *, Raymond Kim, Nikki Mills, Colin Barber, Michel Neeff Department of Paediatric Otorhinolaryngology, Starship Children’s Hospital, 2 Park Road, Grafton, Auckland 1023, New Zealand
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 June 2014 Received in revised form 19 October 2014 Accepted 21 October 2014 Available online xxx
Objectives: This retrospective study reviews the clinical presentation and management of children with airway FBs in our centre. It suggests a safe and reliable guideline to help differentiate which patients should proceed to investigation with rigid laryngobronchoscopy. Methods: A retrospective review of all case notes of laryngobronchoscopies performed for suspected FB aspiration from January 2003 to August 2013 at a tertiary paediatric institution was undertaken. Patient characteristics, history, clinical examination, radiological findings and outcomes were analysed. Results: 158 patients underwent rigid laryngobronchoscopy for suspected FB aspiration between January 2003 and August 2013. The baseline population demographics, the location and type of FBs retrieved were comparable to other similar studies; however, there is a statistically significant higher proportion of Pacific, Maori and Middle Eastern/Latin American/African children compared with the baseline population. Two or more positive findings in the presence of an acute history, any examination or radiology findings is a good indicator to proceed to laryngobronchoscopy with over 99% sensitivity. Conclusion: In a hospital presentation population, this retrospective study suggests that a guideline to proceed to laryngobronchoscopy in a case of suspected FB aspiration is two out of the three positive findings in the presence of an acute history, any examination or radiology findings. Patients who are stable and who do not have two of the three broad category findings can be considered for conservative management and observed on the ward, however, this is a guideline and must be combined with the clinical expertise of the paediatric airway specialist. Further studies are recommended to investigate contributing factors for the disproportionately higher incidence amongst Pacific, Maori and Middle Eastern/Latin American/African children. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Paediatric airway Foreign body aspiration Foreign body inhalation Laryngobronchoscopy Laryngoscopy Bronchoscopy
1. Introduction Possible FB aspiration is a common cause for presentation to the emergency department in the paediatric population. It is one of the leading causes of accidental death in the under one year old age group [1,2]. While it is potentially life threatening, the majority of cases will present in a stable manner and often with non-specific findings [3–7]. Timely diagnosis and management is imperative as delayed diagnosis can result in death as well as complications such as pneumonia, atelectasis, lung abscess or bronchiectasis [8,9]. Rigid laryngobronchoscopy is the most reliable tool for removal of airway FBs [4,10] and is required if there is sufficient suspicion of FB aspiration. However, more commonly the presentation and diagnosis of FB aspiration can be challenging: there may not be a witnessed event suspicious for FB aspiration, the classic triad of cough, wheeze and unilateral decreased breath sounds is
* Corresponding author. Tel.: +64 21 2580407. E-mail address: [email protected] (D. Blair).
frequently absent [8], and a chest X-ray may be normal. Coexisting respiratory pathology may further confuse the clinical picture. This retrospective study reviews the clinical presentation and management of children with airway FBs in our centre. It outlines a safe and reliable way to help differentiate which patients should proceed to investigation with rigid laryngobronchoscopy, and which can be managed conservatively. 2. Methods This study (A + 6084) was approved by the Auckland DHB Research Review Committee (ADHB-RRC) on 27th September 2013. All procedures coded as laryngoscopy and/or bronchoscopy at Starship Children’s Hospital (Auckland, New Zealand) from January 2003 to August 2013 were pooled and patients who underwent this procedure for suspicion of FB aspiration were included. Patient data was collected using the computerised hospital database, which records all procedures performed and all clinical
http://dx.doi.org/10.1016/j.ijporl.2014.10.029 0165-5876/ß 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
2
notes. Baseline characteristics (gender, age, ethnicity), presenting symptoms, clinical examination findings, radiology reports, procedure findings and postoperative course were collected and analysed. If the details being analysed were not recorded in the patient history, the symptom or finding was recorded as negative. Seven patients were excluded from the study as they did not undergo any radiological investigations prior their laryngobronchoscopy. Statistical analysis was carried out using statistics software ‘‘Medcalc’’ and the Chi-squared test was used to compare proportions. 3. Results 3.1. Patient demographics A total of 165 children from January 2003 to August 2013 were identified to have undergone a laryngobronchoscopy with a suspicion of FB aspiration. Seven patients were excluded from the study as they did not undergo any radiological investigations prior their laryngobronchoscopy. Referrals were mainly from the Auckland region but did include referrals from the rest of New Zealand as well as two International referrals from Samoa. 41 (25.9%) patients were transferred from centres outside of Auckland, and 4 patients had already undergone a laryngobronchoscopy prior transfer. The children ranged in age between five months and 14 years (median age: one year nine months). Three-quarters (75%) of the patients were under three years old (Table 1) with a 63.5% male predominance (Table 2). When compared to the Auckland population, there is an over-representation of the Pacific, Maori and Middle Eastern/Latin American/African ethnic groups (Table 3), which are statistically significant (P = 0.0427, 0.0230 and 0.009 respectively), even when adjusted for the Auckland subset of the study population. 3.2. Inhalation to procedure Two-thirds (65.9%) of the patients underwent a laryngobronchoscopy within three days of symptom onset or suspected
Table 2 Gender distribution of study population. Gender
Study population (n)
Study population (%)
Male Female
100 58
63.3 36.7
Total
158
100.0
FB aspiration and 15.4% from greater than three days to one week. In 9.5%, there was a delay of over a month. In 8.9%, the onset of symptoms until procedure time was between one week and one month (Table 4). A FB is more likely to be found in patients with a longer onset of symptoms to procedure time (Table 4). FBs were retrieved in all patients who presented over a month after the onset of symptoms compared with 62.5% in those who presented within the first 24 h. Patients presenting earlier were more likely to present with an acute history of FB aspiration. All patients with three weeks to one month between onset of symptoms to laryngobronchoscopy had positive clinical examination findings. 3.3. Clinical and radiology findings Of the patients with a FB in the airway, 85.2% presented with an acute history of FB aspiration defined as witnessed aspiration, choking or coughing fit (Table 5). Over 90% (91.3%) of patients had positive examination findings which included unilateral diminished breath sounds, wheeze, stridor, increased work of breathing, decreased oxygen saturation or crackles. Almost half (47.0%) had audible wheezing on examination, one-third (33.9%) had unilateral diminished breath sounds, and only 10.4% had stridor (Table 5). 59.1% of the patients had other findings which included increased work of breathing, decreased oxygen saturation and crackles. Less than one in five patients (18.3%) had normal radiology findings. Interestingly, almost half (47.8%) were noted to have air trapping or hyperexpansion seen in their radiology results. In onefifth of patients, a radio-opaque FB was seen. Based on these results, performing a laryngobronchoscopy on one positive finding will identify all patients with a FB in the
Table 1 Age distribution of study population.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx Table 3 Ethnicity of the study population and the Auckland subset of the study population compared with the Auckland population. Auckland Study Auckland Study population population subset of study population 0–14 yrs (n) (%) population (%) 0–14 yrs (%)
Ethnicity
European Maori Pacific people Asian/Indian Middle Eastern/Latin American/African
74 27 32 20 5
46.8 17.1 20.3 12.7 3.2
43.6 16.2 20.5 15.4 4.2
57.1 10.7 14.0 16.7 1.4
airway (sensitivity = 100%) but with a low specificity (2.3%) thus exposing more patients to a laryngobronchoscopy with higher incidence of false positives (Table 6). Any two of the three findings will pick up 99.1% of patients who have aspirated a FB, and increases the specificity ten-fold. Increasing the criteria to three positive findings before proceeding to laryngobronchoscopy will increase the likelihood of correctly diagnosing a patient who has not aspirated a FB, but will also miss the diagnosis in two out of three patients (sensitivity = 59.1%).
3
Table 5 History, examination and radiology findings in patients where a FB was identified in the airway. Patients with FB found in the airway
Number (n)
History Witnessed aspiration Choking or coughing fit Increased work of breathing or persistent cough or wheeze >24 h An acute history (witnessed aspiration or acute choking/coughing fit) Examination findings Unilateral diminished breath sounds Wheeze Stridor Other findings (incl increased WOB, crackles, desaturations) Any positive exam finding Radiology findings Radio-opaque FB Air trapping/hyperexpansion Atelectasis/consolidation Other e.g. mediastinal shift, tracheal deviation, peribronchial thickening Any positive radiology finding
%
85 94 63
73.9 81.7 54.8
98
85.2
39 54 11 68
33.9 47.0 9.6 59.1
105
91.3
24 55 25 15
20.9 47.8 21.7 13.0
94
81.7
3.4. Complications Two patients (1.7%) returned with chest infections after being discharged. Eight patients (7.0%) underwent a repeat laryngobronchoscopy. Two of these were for removal of FB fragments that were unable to be removed at the initial procedure. In two patients, this was due to inability to further instrument in the first procedure due to swelling. In another two, there was difficulty assessing whether all FBs were removed at the initial procedure. In the last two, it was performed to investigate unresolved symptoms but no retained FBs were identified. Four (3.5%) patients had undergone an earlier laryngobronchoscopy which had failed to identify a FB. All of these cases were performed at other centres. In one patient, a laryngobronchoscopy identifying a FB was performed at the referral hospital but it was unable to be retrieved. This patient underwent an emergency tracheostomy prior to transfer. Two patients presented with two out of three of the criteria but were discharged initially. When they re-presented to hospital, a FB was found on laryngobronchoscopy. One patient from our centre was missed as they did not have a history of FB aspiration. This patient was seen by the medical team and discharged with a diagnosis of croup. The other patient was initially seen at a peripheral hospital. This patient had a suspicious history of FB aspiration and was wheezy on examination. He was discharged when the symptoms improved with treatment but subsequently re-presented. Two patients were referred from Samoa with two and three positive findings. They had been treated conservatively for two weeks with IV antibiotics without resolution of symptoms and then transferred for surgical management.
Table 4 n (%) Onset of symptom to procedure <1 day 1–3 days >3–7 days >1 week–1 month >1 month Total
72 32 24 14
(45.6) (20.3) (15.2) (8.9)
15 (9.5) 158
FB found (%) 45 24 20 11
(62.5) 45 (100.0) (75) 24 (100.0) (83.3) 15 (66.7) (78.6) 6 (54.5)
15 (100) 115
FB found FB found and +ve and +ve acute hx (%) exam (%)
8 (53.3) 98
37 23 20 11
(82.2) (95.8) (100.0) (100.0)
14 (93.3) 105
FB found and +ve rad (%) 37 18 15 11
(82.2) (75.0) (75.0) (100.0)
13 (86.7) 94
One patient underwent removal of FB via an external approach. This patient had aspirated a metal bristle of a barbeque steel brush. He presented with a three-month history of intermittent stridor and hoarseness. His X-ray showed a thin line along the aryepiglottic fold in keeping with a FB. A FB was not identified at his initial laryngobronchoscopy but a swelling over the right vocal fold was seen and biopsied. The intraoperative findings were discussed at the Radiology conference meeting and a plan was made to discharge the patient with a MRI pending the biopsy results. The biopsy confirmed granulation tissue and the patient was recalled to hospital. A repeat X-ray showed the linear FB had migrated to the soft tissues of the chin, anterior to the hyoid from where it was removed surgically. 4. Discussion 4.1. Patient demographics The overall background demographics of the patients are inline with similar series [4,5,11–13]. The male to female ratio was 1.7:1. This is a well-documented finding [11] and is thought to be due to the more physical play in male children [14]. In our study population, 74.5% of the patients were under 3 years, which is also in keeping with other series [4,5,11–13]. There is a significant overrepresentation of Pacific, Maori and Middle Eastern/Latin American/ African ethnicities by 6.3%, 5.5% and 2.8% respectively (P = 0.0427, 0.0230 and 0.0099) compared to the Auckland population [14]. This difference is observed even after adjustment to include the Auckland subset of the study population only (Table 3). The reason for this is unclear but a similar sized 10-year retrospective study from Australia linked patients from non-English speaking backgrounds with a statistically significant higher risk of food aspiration especially nut aspiration in Arabic-speaking households [15]. This Table 6
1. Positive findinga 2. Positive findingsa 3. Positive findingsa
Sensitivity
Specificity
PPV
NPV
100.0% 99.1% 59.1%
2.3% 23.3% 74.4%
73.3% 77.6% 86.1%
100.0% 90.9% 40.5%
a Positive finding = acute history, any examination finding, or any abnormality of radiology.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
4
Any positive finding in two of the three following categories: 1. Acute history • •
Witnessed aspiration Acute choking or coughing episode
2. Examination • • • • • •
Unilateral diminished breath sounds Wheeze Stridor Crackles Increased work of breathing Desaturations
3. Radiology • • • • •
Radio-opaque FB Air trapping or hyperexpansion Mediastinal shift Atelectasis Consolidation
Fig. 1. Heuristic criteria: consider laryngobronchosopy if there are positive findings in any two categories.
is speculated to be largely due to cultural influences in the types of foods given to children. Further studies are recommended to investigate reasons for the Pacific and Maori ethnic over-representation in this study. 4.2. Clinical and radiological findings The diagnosis of FB aspiration can often be challenging. While some patients will present in acute respiratory distress requiring urgent laryngobronchoscopy to retrieve the FB, more commonly, patients present in a stable condition. To add to the uncertainty, the event may have been unwitnessed and examination of the patient often may not reveal any specific findings. There are a number of studies which have looked at the sensitivity and specificity of the history, examination and radiological findings in the diagnostic evaluation of suspected FB aspiration
[4,5,11,13,16,17]. A review by Fidkowski et al. [11] totaling over 12,000 patients reported varying sensitivity and specificity of witnessed histories of FB aspiration ranging from 58.3% [18] to 93.2% [19] and 32.1% [20] to 96.3% [21] respectively. While the classic triad of sudden onset cough, wheeze and unilateral diminished breath sounds is the textbook description of the presentation of FB aspiration [8], it is present in less than 40% of patients [16] as also confirmed in this review (7.8%). Moreover, clinical signs and X-ray findings either lack sensitivity and/or specificity [11]. Of the patients who had a FB identified in this study, 18.3% had normal X-rays. One-fifth (20.9%) had radioopaque FBs seen on radiology and less than half (47.8%) had air trapping or hyperexpansion on radiology. In cases of suspected FB aspiration, the accepted management is to proceed to laryngobronchoscopy due to the potentially serious complications of a delayed diagnosis. Studies suggest a low index of suspicion to proceed to laryngobronchoscopy [4,13,16,17] have not specified parameters for what this entails, while other studies recommend laryngobronchoscopy on an appropriate history alone [5]. Heyer et al. [7] suggest that the presence of a witnessed choking history, focal hyperinflation on chest X-ray and increased white blood cell count was a strong indicator to proceed to a laryngobronchoscopy as a FB was recovered in almost all these cases [7]. There were no comments on how many diagnoses would have been missed (i.e. false negative) using this method alone. We found that two or more positive findings in the history, examination or radiology are a good indicator to proceed to laryngobronchoscopy (Fig. 1). Using this guide, all but one (99.1%) of the patients with an airway FB were identified. Applying this to our study population would have decreased the rate of negative laryngobronchoscopies to 22.4% as compared with our current rate of 27.2%. There is no demonstrable difference (P-value = 0.148) but a further study to look at the efficacy of applying this is recommended. The patient who was missed was a 17-month-old girl who did not have an acute history and also a normal X-ray. She presented with a six-week history of wheeze resistant to medical treatment. The reported rate of negative laryngobronchoscopies for suspected FB aspiration varies from 28.7% [4] to 43% [13]. When comparing to other similar studies, the negative
Fig. 2. Decision tree in the management of a child with suspected foreign body aspiration.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
laryngobronchoscopy rate for FB aspiration in our centre tends towards the lower end. The reason for this is uncertain but may be contributed by this algorithm being increasingly adopted over the past three years. 4.3. A heuristic approach to FBs in the paediatric airway From the data of this retrospective study, we have developed the below approach (Fig. 2) in the management of suspected FB aspiration. In the absence of any acute history findings, clinicians should also bear in mind the possibility of FB aspiration in their differential diagnoses as the symptoms can masquerade in a variety of presentations. As such, unusual presentations or failure to follow a projected recovery course despite adequate treatment may warrant definitive investigation with laryngobronchoscopy. 5. Conclusion 158 patients underwent a laryngobronchoscopy for suspected FB aspiration between January 2003 and August 2013. 115 FBs were identified and all were retrieved. There were no associated deaths in our study population. The baseline population demographics, the location and the type of FBs retrieved were comparable to other similar studies; however, there is a statistically significant higher proportion of Maori and Pacific children compared with the baseline population. Further studies are recommended to investigate contributing factors to this finding. This retrospective study suggests that a guideline to proceed to laryngobronchoscopy in a case of suspected FB aspiration is two or more positive findings in the presence of an acute history, any clinical examination findings or any radiology findings. We foresee this simplified diagnostic algorithm to be applicable in a wide variety of settings, assisting clinicians to quickly identify those who should undergo further formal investigation. Due to the retrospective nature of this analysis, this algorithm should be regarded as a guideline and a low threshold to consult a paediatric airway specialist is recommended. Acknowledgement We wish to thank Joel Tham (B.Com/B.Sc) for his assistance in the statistic analysis of this manuscript.
5
References [1] B.D. Lifschultz, E.R. Donoghue, Deaths due to FB aspiration in children: the continuing hazard of toy balloons, J. Forensic Sci. 41 (March (2)) (1996) 247–251. [2] Child and Youth Mortality Review Committee New Zealand, Special Report: Unintentional Suffocation, FB Inhalation and Strangulation, Health Quality & Safety Commission, Wellington, New Zealand, March 2013. [3] O. Higuchi, Y. Adachi, T. Ichimaru, M. Asai, K. Kawasaki, FB aspiration in children: a nationwide survey in Japan, Int. J. Pediatr. Otorhinolaryngol. 73 (May (5)) (2009) 659–661. [4] M. Tomaske, A.C. Gerber, S. Stocker, M. Weiss, Tracheobronchial foreign body aspiration in children – diagnostic value of symptoms and signs, Swiss Med. Wkly. 136 (19 August (33–34)) (2006) 533–538. [5] B. Tokar, R. Ozkan, H. Ilhan, Tracheobronchial foreign bodies in children: importance of accurate history and plain chest radiography in delayed presentation, Clin. Radiol. 59 (July (7)) (2004) 609–615. [6] G. Kiyan, B. Gocmen, H. Tugtepe, F. Karakoc, E. Dagli, T.E. Dagli, FB aspiration in children: the value of diagnostic criteria, Int. J. Pediatr. Otorhinolaryngol. 73 (July (7)) (2009) 963–967. [7] C.M. Heyer, M.E. Bollmeier, L. Rossler, T.G. Nuesslein, V. Stephan, T.T. Bauer, et al., Evaluation of clinical, radiological and laboratory prebronchoscopy findings in children with suspected FB aspiration, J. Pediatr. Surg. 41 (2006) 1882–1888. [8] H. Tan, K. Brown, T. McGill, M.A. Kenna, D.P. Lund, G.B. Healy, Airway foreign bodies (FB): a 10-year review, Int. J. Pediatr. Otorhinolaryngol. 56 (1 December (2)) (2000) 91–99. [9] W. Gang, P. Zhengxia, L. Hongbo, L. Yonggang, D. Jiangtao, W. Shengde, et al., Diagnosis and treatment of tracheobronchial foreign bodies in 1024 children, J. Pediatr. Surg. 47 (November (11)) (2012) 2004–2010. [10] W. Korlacki, K. Korecka, J. Dzielicki, Foreign body aspiration in children: diagnostic and therapeutic role of bronchoscopy, Pediatr. Surg. Int. 27 (August (8)) (2011) 833–837. [11] C.W. Fidkowski, H. Zheng, P.G. Firth, The anesthetic considerations of tracheobronchial foreign bodies in children: a literature review of 12,979 cases, Anesth. Analg. 111 (October (4)) (2010) 1016–1025. [12] N. Mani, M. Soma, S. Massey, D. Albert, C.M. Bailey, Removal of inhaled foreign bodies – middle of the night or the next morning? Int. J. Pediatr. Otorhinolaryngol. 73 (August (8)) (2009) 1085–1089. [13] E. Lea, H. Nawaf, T. Yoav, S. Elvin, Z. Ze’ev, K. Amir, Diagnostic evaluation of foreign body aspiration in children: a prospective study, J. Pediatr. Surg. 40 (2005) 1122–1127. [14] K. Mantel, I. Butenandt, Tracheobronchial foreign body aspiration in childhood. A report on 224 cases, Eur. J. Paediatr. 145 (August (3)) (1986) 211–216. [15] Statistics New Zealand, Census of Population and Dwellings, Statistics New Zealand, Wellington, 2006, Data retrieved from www.stats.govt.nz/Census.aspx. [16] S. Choroomi, J. Curotta, FB aspiration and language spoken at home: 10-year review, J. Laryngol. Otol. 125 (July (7)) (2011) 719–723. [17] A.M. Shubha, K. Das, Tracheobronchial foreign bodies in infants, Int. J. Pediatr. Otorhinolaryngol. 73 (October (10)) (2009) 1385–1389. [18] F. Og˘uz, A. Citak, E. Unu¨var, M. Sidal, Airway foreign bodies in childhood, Int. J. Pediatr. Otorhinolaryngol. 52 (30 January (1)) (2000) 11–16. [19] V. Erikc¸i, S. Karac¸ay, A. Arikan, Foreign body aspiration: a four-years experience, Ulus. Travma Acil Cerrahi Derg. 9 (2003) 45–49. [20] L.B. Aydogan, U. Tuncer, L. Soylu, M. Kiroglu, C. Ozsahinoglu, Rigid bronchoscopy for the suspicion of foreign body in the airway, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 823–828. [21] S.R. Cohen, K.A. Geller, Anesthesia and pediatric endoscopy: the surgeon’s view, Otolaryngol. Clin. N. Am. 14 (1981) 705–713.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
PEDOT-7336; No. of Pages 5 International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
A heuristic approach to foreign bodies in the paediatric airway Dora Blair *, Raymond Kim, Nikki Mills, Colin Barber, Michel Neeff Department of Paediatric Otorhinolaryngology, Starship Children’s Hospital, 2 Park Road, Grafton, Auckland 1023, New Zealand
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 June 2014 Received in revised form 19 October 2014 Accepted 21 October 2014 Available online xxx
Objectives: This retrospective study reviews the clinical presentation and management of children with airway FBs in our centre. It suggests a safe and reliable guideline to help differentiate which patients should proceed to investigation with rigid laryngobronchoscopy. Methods: A retrospective review of all case notes of laryngobronchoscopies performed for suspected FB aspiration from January 2003 to August 2013 at a tertiary paediatric institution was undertaken. Patient characteristics, history, clinical examination, radiological findings and outcomes were analysed. Results: 158 patients underwent rigid laryngobronchoscopy for suspected FB aspiration between January 2003 and August 2013. The baseline population demographics, the location and type of FBs retrieved were comparable to other similar studies; however, there is a statistically significant higher proportion of Pacific, Maori and Middle Eastern/Latin American/African children compared with the baseline population. Two or more positive findings in the presence of an acute history, any examination or radiology findings is a good indicator to proceed to laryngobronchoscopy with over 99% sensitivity. Conclusion: In a hospital presentation population, this retrospective study suggests that a guideline to proceed to laryngobronchoscopy in a case of suspected FB aspiration is two out of the three positive findings in the presence of an acute history, any examination or radiology findings. Patients who are stable and who do not have two of the three broad category findings can be considered for conservative management and observed on the ward, however, this is a guideline and must be combined with the clinical expertise of the paediatric airway specialist. Further studies are recommended to investigate contributing factors for the disproportionately higher incidence amongst Pacific, Maori and Middle Eastern/Latin American/African children. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Paediatric airway Foreign body aspiration Foreign body inhalation Laryngobronchoscopy Laryngoscopy Bronchoscopy
1. Introduction Possible FB aspiration is a common cause for presentation to the emergency department in the paediatric population. It is one of the leading causes of accidental death in the under one year old age group [1,2]. While it is potentially life threatening, the majority of cases will present in a stable manner and often with non-specific findings [3–7]. Timely diagnosis and management is imperative as delayed diagnosis can result in death as well as complications such as pneumonia, atelectasis, lung abscess or bronchiectasis [8,9]. Rigid laryngobronchoscopy is the most reliable tool for removal of airway FBs [4,10] and is required if there is sufficient suspicion of FB aspiration. However, more commonly the presentation and diagnosis of FB aspiration can be challenging: there may not be a witnessed event suspicious for FB aspiration, the classic triad of cough, wheeze and unilateral decreased breath sounds is
* Corresponding author. Tel.: +64 21 2580407. E-mail address: [email protected] (D. Blair).
frequently absent [8], and a chest X-ray may be normal. Coexisting respiratory pathology may further confuse the clinical picture. This retrospective study reviews the clinical presentation and management of children with airway FBs in our centre. It outlines a safe and reliable way to help differentiate which patients should proceed to investigation with rigid laryngobronchoscopy, and which can be managed conservatively. 2. Methods This study (A + 6084) was approved by the Auckland DHB Research Review Committee (ADHB-RRC) on 27th September 2013. All procedures coded as laryngoscopy and/or bronchoscopy at Starship Children’s Hospital (Auckland, New Zealand) from January 2003 to August 2013 were pooled and patients who underwent this procedure for suspicion of FB aspiration were included. Patient data was collected using the computerised hospital database, which records all procedures performed and all clinical
http://dx.doi.org/10.1016/j.ijporl.2014.10.029 0165-5876/ß 2014 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
2
notes. Baseline characteristics (gender, age, ethnicity), presenting symptoms, clinical examination findings, radiology reports, procedure findings and postoperative course were collected and analysed. If the details being analysed were not recorded in the patient history, the symptom or finding was recorded as negative. Seven patients were excluded from the study as they did not undergo any radiological investigations prior their laryngobronchoscopy. Statistical analysis was carried out using statistics software ‘‘Medcalc’’ and the Chi-squared test was used to compare proportions. 3. Results 3.1. Patient demographics A total of 165 children from January 2003 to August 2013 were identified to have undergone a laryngobronchoscopy with a suspicion of FB aspiration. Seven patients were excluded from the study as they did not undergo any radiological investigations prior their laryngobronchoscopy. Referrals were mainly from the Auckland region but did include referrals from the rest of New Zealand as well as two International referrals from Samoa. 41 (25.9%) patients were transferred from centres outside of Auckland, and 4 patients had already undergone a laryngobronchoscopy prior transfer. The children ranged in age between five months and 14 years (median age: one year nine months). Three-quarters (75%) of the patients were under three years old (Table 1) with a 63.5% male predominance (Table 2). When compared to the Auckland population, there is an over-representation of the Pacific, Maori and Middle Eastern/Latin American/African ethnic groups (Table 3), which are statistically significant (P = 0.0427, 0.0230 and 0.009 respectively), even when adjusted for the Auckland subset of the study population. 3.2. Inhalation to procedure Two-thirds (65.9%) of the patients underwent a laryngobronchoscopy within three days of symptom onset or suspected
Table 2 Gender distribution of study population. Gender
Study population (n)
Study population (%)
Male Female
100 58
63.3 36.7
Total
158
100.0
FB aspiration and 15.4% from greater than three days to one week. In 9.5%, there was a delay of over a month. In 8.9%, the onset of symptoms until procedure time was between one week and one month (Table 4). A FB is more likely to be found in patients with a longer onset of symptoms to procedure time (Table 4). FBs were retrieved in all patients who presented over a month after the onset of symptoms compared with 62.5% in those who presented within the first 24 h. Patients presenting earlier were more likely to present with an acute history of FB aspiration. All patients with three weeks to one month between onset of symptoms to laryngobronchoscopy had positive clinical examination findings. 3.3. Clinical and radiology findings Of the patients with a FB in the airway, 85.2% presented with an acute history of FB aspiration defined as witnessed aspiration, choking or coughing fit (Table 5). Over 90% (91.3%) of patients had positive examination findings which included unilateral diminished breath sounds, wheeze, stridor, increased work of breathing, decreased oxygen saturation or crackles. Almost half (47.0%) had audible wheezing on examination, one-third (33.9%) had unilateral diminished breath sounds, and only 10.4% had stridor (Table 5). 59.1% of the patients had other findings which included increased work of breathing, decreased oxygen saturation and crackles. Less than one in five patients (18.3%) had normal radiology findings. Interestingly, almost half (47.8%) were noted to have air trapping or hyperexpansion seen in their radiology results. In onefifth of patients, a radio-opaque FB was seen. Based on these results, performing a laryngobronchoscopy on one positive finding will identify all patients with a FB in the
Table 1 Age distribution of study population.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx Table 3 Ethnicity of the study population and the Auckland subset of the study population compared with the Auckland population. Auckland Study Auckland Study population population subset of study population 0–14 yrs (n) (%) population (%) 0–14 yrs (%)
Ethnicity
European Maori Pacific people Asian/Indian Middle Eastern/Latin American/African
74 27 32 20 5
46.8 17.1 20.3 12.7 3.2
43.6 16.2 20.5 15.4 4.2
57.1 10.7 14.0 16.7 1.4
airway (sensitivity = 100%) but with a low specificity (2.3%) thus exposing more patients to a laryngobronchoscopy with higher incidence of false positives (Table 6). Any two of the three findings will pick up 99.1% of patients who have aspirated a FB, and increases the specificity ten-fold. Increasing the criteria to three positive findings before proceeding to laryngobronchoscopy will increase the likelihood of correctly diagnosing a patient who has not aspirated a FB, but will also miss the diagnosis in two out of three patients (sensitivity = 59.1%).
3
Table 5 History, examination and radiology findings in patients where a FB was identified in the airway. Patients with FB found in the airway
Number (n)
History Witnessed aspiration Choking or coughing fit Increased work of breathing or persistent cough or wheeze >24 h An acute history (witnessed aspiration or acute choking/coughing fit) Examination findings Unilateral diminished breath sounds Wheeze Stridor Other findings (incl increased WOB, crackles, desaturations) Any positive exam finding Radiology findings Radio-opaque FB Air trapping/hyperexpansion Atelectasis/consolidation Other e.g. mediastinal shift, tracheal deviation, peribronchial thickening Any positive radiology finding
%
85 94 63
73.9 81.7 54.8
98
85.2
39 54 11 68
33.9 47.0 9.6 59.1
105
91.3
24 55 25 15
20.9 47.8 21.7 13.0
94
81.7
3.4. Complications Two patients (1.7%) returned with chest infections after being discharged. Eight patients (7.0%) underwent a repeat laryngobronchoscopy. Two of these were for removal of FB fragments that were unable to be removed at the initial procedure. In two patients, this was due to inability to further instrument in the first procedure due to swelling. In another two, there was difficulty assessing whether all FBs were removed at the initial procedure. In the last two, it was performed to investigate unresolved symptoms but no retained FBs were identified. Four (3.5%) patients had undergone an earlier laryngobronchoscopy which had failed to identify a FB. All of these cases were performed at other centres. In one patient, a laryngobronchoscopy identifying a FB was performed at the referral hospital but it was unable to be retrieved. This patient underwent an emergency tracheostomy prior to transfer. Two patients presented with two out of three of the criteria but were discharged initially. When they re-presented to hospital, a FB was found on laryngobronchoscopy. One patient from our centre was missed as they did not have a history of FB aspiration. This patient was seen by the medical team and discharged with a diagnosis of croup. The other patient was initially seen at a peripheral hospital. This patient had a suspicious history of FB aspiration and was wheezy on examination. He was discharged when the symptoms improved with treatment but subsequently re-presented. Two patients were referred from Samoa with two and three positive findings. They had been treated conservatively for two weeks with IV antibiotics without resolution of symptoms and then transferred for surgical management.
Table 4 n (%) Onset of symptom to procedure <1 day 1–3 days >3–7 days >1 week–1 month >1 month Total
72 32 24 14
(45.6) (20.3) (15.2) (8.9)
15 (9.5) 158
FB found (%) 45 24 20 11
(62.5) 45 (100.0) (75) 24 (100.0) (83.3) 15 (66.7) (78.6) 6 (54.5)
15 (100) 115
FB found FB found and +ve and +ve acute hx (%) exam (%)
8 (53.3) 98
37 23 20 11
(82.2) (95.8) (100.0) (100.0)
14 (93.3) 105
FB found and +ve rad (%) 37 18 15 11
(82.2) (75.0) (75.0) (100.0)
13 (86.7) 94
One patient underwent removal of FB via an external approach. This patient had aspirated a metal bristle of a barbeque steel brush. He presented with a three-month history of intermittent stridor and hoarseness. His X-ray showed a thin line along the aryepiglottic fold in keeping with a FB. A FB was not identified at his initial laryngobronchoscopy but a swelling over the right vocal fold was seen and biopsied. The intraoperative findings were discussed at the Radiology conference meeting and a plan was made to discharge the patient with a MRI pending the biopsy results. The biopsy confirmed granulation tissue and the patient was recalled to hospital. A repeat X-ray showed the linear FB had migrated to the soft tissues of the chin, anterior to the hyoid from where it was removed surgically. 4. Discussion 4.1. Patient demographics The overall background demographics of the patients are inline with similar series [4,5,11–13]. The male to female ratio was 1.7:1. This is a well-documented finding [11] and is thought to be due to the more physical play in male children [14]. In our study population, 74.5% of the patients were under 3 years, which is also in keeping with other series [4,5,11–13]. There is a significant overrepresentation of Pacific, Maori and Middle Eastern/Latin American/ African ethnicities by 6.3%, 5.5% and 2.8% respectively (P = 0.0427, 0.0230 and 0.0099) compared to the Auckland population [14]. This difference is observed even after adjustment to include the Auckland subset of the study population only (Table 3). The reason for this is unclear but a similar sized 10-year retrospective study from Australia linked patients from non-English speaking backgrounds with a statistically significant higher risk of food aspiration especially nut aspiration in Arabic-speaking households [15]. This Table 6
1. Positive findinga 2. Positive findingsa 3. Positive findingsa
Sensitivity
Specificity
PPV
NPV
100.0% 99.1% 59.1%
2.3% 23.3% 74.4%
73.3% 77.6% 86.1%
100.0% 90.9% 40.5%
a Positive finding = acute history, any examination finding, or any abnormality of radiology.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
4
Any positive finding in two of the three following categories: 1. Acute history • •
Witnessed aspiration Acute choking or coughing episode
2. Examination • • • • • •
Unilateral diminished breath sounds Wheeze Stridor Crackles Increased work of breathing Desaturations
3. Radiology • • • • •
Radio-opaque FB Air trapping or hyperexpansion Mediastinal shift Atelectasis Consolidation
Fig. 1. Heuristic criteria: consider laryngobronchosopy if there are positive findings in any two categories.
is speculated to be largely due to cultural influences in the types of foods given to children. Further studies are recommended to investigate reasons for the Pacific and Maori ethnic over-representation in this study. 4.2. Clinical and radiological findings The diagnosis of FB aspiration can often be challenging. While some patients will present in acute respiratory distress requiring urgent laryngobronchoscopy to retrieve the FB, more commonly, patients present in a stable condition. To add to the uncertainty, the event may have been unwitnessed and examination of the patient often may not reveal any specific findings. There are a number of studies which have looked at the sensitivity and specificity of the history, examination and radiological findings in the diagnostic evaluation of suspected FB aspiration
[4,5,11,13,16,17]. A review by Fidkowski et al. [11] totaling over 12,000 patients reported varying sensitivity and specificity of witnessed histories of FB aspiration ranging from 58.3% [18] to 93.2% [19] and 32.1% [20] to 96.3% [21] respectively. While the classic triad of sudden onset cough, wheeze and unilateral diminished breath sounds is the textbook description of the presentation of FB aspiration [8], it is present in less than 40% of patients [16] as also confirmed in this review (7.8%). Moreover, clinical signs and X-ray findings either lack sensitivity and/or specificity [11]. Of the patients who had a FB identified in this study, 18.3% had normal X-rays. One-fifth (20.9%) had radioopaque FBs seen on radiology and less than half (47.8%) had air trapping or hyperexpansion on radiology. In cases of suspected FB aspiration, the accepted management is to proceed to laryngobronchoscopy due to the potentially serious complications of a delayed diagnosis. Studies suggest a low index of suspicion to proceed to laryngobronchoscopy [4,13,16,17] have not specified parameters for what this entails, while other studies recommend laryngobronchoscopy on an appropriate history alone [5]. Heyer et al. [7] suggest that the presence of a witnessed choking history, focal hyperinflation on chest X-ray and increased white blood cell count was a strong indicator to proceed to a laryngobronchoscopy as a FB was recovered in almost all these cases [7]. There were no comments on how many diagnoses would have been missed (i.e. false negative) using this method alone. We found that two or more positive findings in the history, examination or radiology are a good indicator to proceed to laryngobronchoscopy (Fig. 1). Using this guide, all but one (99.1%) of the patients with an airway FB were identified. Applying this to our study population would have decreased the rate of negative laryngobronchoscopies to 22.4% as compared with our current rate of 27.2%. There is no demonstrable difference (P-value = 0.148) but a further study to look at the efficacy of applying this is recommended. The patient who was missed was a 17-month-old girl who did not have an acute history and also a normal X-ray. She presented with a six-week history of wheeze resistant to medical treatment. The reported rate of negative laryngobronchoscopies for suspected FB aspiration varies from 28.7% [4] to 43% [13]. When comparing to other similar studies, the negative
Fig. 2. Decision tree in the management of a child with suspected foreign body aspiration.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029
G Model
PEDOT-7336; No. of Pages 5 D. Blair et al. / International Journal of Pediatric Otorhinolaryngology xxx (2014) xxx–xxx
laryngobronchoscopy rate for FB aspiration in our centre tends towards the lower end. The reason for this is uncertain but may be contributed by this algorithm being increasingly adopted over the past three years. 4.3. A heuristic approach to FBs in the paediatric airway From the data of this retrospective study, we have developed the below approach (Fig. 2) in the management of suspected FB aspiration. In the absence of any acute history findings, clinicians should also bear in mind the possibility of FB aspiration in their differential diagnoses as the symptoms can masquerade in a variety of presentations. As such, unusual presentations or failure to follow a projected recovery course despite adequate treatment may warrant definitive investigation with laryngobronchoscopy. 5. Conclusion 158 patients underwent a laryngobronchoscopy for suspected FB aspiration between January 2003 and August 2013. 115 FBs were identified and all were retrieved. There were no associated deaths in our study population. The baseline population demographics, the location and the type of FBs retrieved were comparable to other similar studies; however, there is a statistically significant higher proportion of Maori and Pacific children compared with the baseline population. Further studies are recommended to investigate contributing factors to this finding. This retrospective study suggests that a guideline to proceed to laryngobronchoscopy in a case of suspected FB aspiration is two or more positive findings in the presence of an acute history, any clinical examination findings or any radiology findings. We foresee this simplified diagnostic algorithm to be applicable in a wide variety of settings, assisting clinicians to quickly identify those who should undergo further formal investigation. Due to the retrospective nature of this analysis, this algorithm should be regarded as a guideline and a low threshold to consult a paediatric airway specialist is recommended. Acknowledgement We wish to thank Joel Tham (B.Com/B.Sc) for his assistance in the statistic analysis of this manuscript.
5
References [1] B.D. Lifschultz, E.R. Donoghue, Deaths due to FB aspiration in children: the continuing hazard of toy balloons, J. Forensic Sci. 41 (March (2)) (1996) 247–251. [2] Child and Youth Mortality Review Committee New Zealand, Special Report: Unintentional Suffocation, FB Inhalation and Strangulation, Health Quality & Safety Commission, Wellington, New Zealand, March 2013. [3] O. Higuchi, Y. Adachi, T. Ichimaru, M. Asai, K. Kawasaki, FB aspiration in children: a nationwide survey in Japan, Int. J. Pediatr. Otorhinolaryngol. 73 (May (5)) (2009) 659–661. [4] M. Tomaske, A.C. Gerber, S. Stocker, M. Weiss, Tracheobronchial foreign body aspiration in children – diagnostic value of symptoms and signs, Swiss Med. Wkly. 136 (19 August (33–34)) (2006) 533–538. [5] B. Tokar, R. Ozkan, H. Ilhan, Tracheobronchial foreign bodies in children: importance of accurate history and plain chest radiography in delayed presentation, Clin. Radiol. 59 (July (7)) (2004) 609–615. [6] G. Kiyan, B. Gocmen, H. Tugtepe, F. Karakoc, E. Dagli, T.E. Dagli, FB aspiration in children: the value of diagnostic criteria, Int. J. Pediatr. Otorhinolaryngol. 73 (July (7)) (2009) 963–967. [7] C.M. Heyer, M.E. Bollmeier, L. Rossler, T.G. Nuesslein, V. Stephan, T.T. Bauer, et al., Evaluation of clinical, radiological and laboratory prebronchoscopy findings in children with suspected FB aspiration, J. Pediatr. Surg. 41 (2006) 1882–1888. [8] H. Tan, K. Brown, T. McGill, M.A. Kenna, D.P. Lund, G.B. Healy, Airway foreign bodies (FB): a 10-year review, Int. J. Pediatr. Otorhinolaryngol. 56 (1 December (2)) (2000) 91–99. [9] W. Gang, P. Zhengxia, L. Hongbo, L. Yonggang, D. Jiangtao, W. Shengde, et al., Diagnosis and treatment of tracheobronchial foreign bodies in 1024 children, J. Pediatr. Surg. 47 (November (11)) (2012) 2004–2010. [10] W. Korlacki, K. Korecka, J. Dzielicki, Foreign body aspiration in children: diagnostic and therapeutic role of bronchoscopy, Pediatr. Surg. Int. 27 (August (8)) (2011) 833–837. [11] C.W. Fidkowski, H. Zheng, P.G. Firth, The anesthetic considerations of tracheobronchial foreign bodies in children: a literature review of 12,979 cases, Anesth. Analg. 111 (October (4)) (2010) 1016–1025. [12] N. Mani, M. Soma, S. Massey, D. Albert, C.M. Bailey, Removal of inhaled foreign bodies – middle of the night or the next morning? Int. J. Pediatr. Otorhinolaryngol. 73 (August (8)) (2009) 1085–1089. [13] E. Lea, H. Nawaf, T. Yoav, S. Elvin, Z. Ze’ev, K. Amir, Diagnostic evaluation of foreign body aspiration in children: a prospective study, J. Pediatr. Surg. 40 (2005) 1122–1127. [14] K. Mantel, I. Butenandt, Tracheobronchial foreign body aspiration in childhood. A report on 224 cases, Eur. J. Paediatr. 145 (August (3)) (1986) 211–216. [15] Statistics New Zealand, Census of Population and Dwellings, Statistics New Zealand, Wellington, 2006, Data retrieved from www.stats.govt.nz/Census.aspx. [16] S. Choroomi, J. Curotta, FB aspiration and language spoken at home: 10-year review, J. Laryngol. Otol. 125 (July (7)) (2011) 719–723. [17] A.M. Shubha, K. Das, Tracheobronchial foreign bodies in infants, Int. J. Pediatr. Otorhinolaryngol. 73 (October (10)) (2009) 1385–1389. [18] F. Og˘uz, A. Citak, E. Unu¨var, M. Sidal, Airway foreign bodies in childhood, Int. J. Pediatr. Otorhinolaryngol. 52 (30 January (1)) (2000) 11–16. [19] V. Erikc¸i, S. Karac¸ay, A. Arikan, Foreign body aspiration: a four-years experience, Ulus. Travma Acil Cerrahi Derg. 9 (2003) 45–49. [20] L.B. Aydogan, U. Tuncer, L. Soylu, M. Kiroglu, C. Ozsahinoglu, Rigid bronchoscopy for the suspicion of foreign body in the airway, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 823–828. [21] S.R. Cohen, K.A. Geller, Anesthesia and pediatric endoscopy: the surgeon’s view, Otolaryngol. Clin. N. Am. 14 (1981) 705–713.
Please cite this article in press as: D. Blair, et al., A heuristic approach to foreign bodies in the paediatric airway, Int. J. Pediatr. Otorhinolaryngol. (2014), http://dx.doi.org/10.1016/j.ijporl.2014.10.029