Determinants of positive rigid bronchoscopy for suspected organic foreign body aspiration in children younger than five years

Abstract

Purpose

Rigid bronchoscopy is widely used for diagnosis and treatment of foreign body aspiration (FBA) in children, but negative results can be reported, especially with radiolucent organic FBA. This study aimed to evaluate the clinical features and pre-procedure predictors of organic FBA in children under 5 years of age.

Methods

Children aged less than 5 years old who underwent rigid bronchoscopy for suspected organic FBA were retrospectively evaluated for demographics, history of aspiration, relevant clinical symptoms and signs, radiological findings, in addition to type and location of foreign body (FB). To determine the predictors of positive FBA, stepwise backward logistic regression was applied.

Results

A total of 228 children were included (69.7% boys). The mean age was 1.98 ± 1.07 years. Foreign bodies were found in 202 cases (88.59%). Age ≤ 3 years and history of witnessed choking had the highest sensitivity while diminished breath sounds had the highest specificity for diagnosing FBA. In multivariate analysis, witnessed choking, wheezy chest, diminished breath sounds, and respiratory distress were independently associated with increased risk of FBA.

Conclusion

Children under 5 years of age with suspected organic FBA have varied clinical and radiological findings. The history of choking, wheezy chest, diminished air entry, or respiratory distress increases the chance of positive rigid bronchoscopy in this age group.

Introduction

Foreign body aspiration (FBA) is a major life-threatening clinical entity in children, accounting for 5% of all accidental deaths in children under 4 years of age [1] and associated with a mortality rate of up to 11% [2]. Young children, especially those under 5 years of age, are at high risk for FBA, accounting for up to 98% of cases [3]. Aspirated foreign bodies (FBs) can be classified as organic or inorganic, radiopaque or radiolucent, with radiolucent organic FBs being the most common. Diagnosing radiolucent FBs can be challenging since children may present with mild or nonspecific symptoms, and radiological findings may be negative [4].

Since its first use by the Otorhinolaryngologist, Gustav Killian, in 1876 for tracheal FB extraction [5], rigid bronchoscopy (RB) has been widely used for the diagnosis and treatment of FBA [6]. However, the negative results of rigid bronchoscopy in children with radiolucent FBA are still reported in literature [7, 8], especially when the clinical and radiological features are insufficient to exclude the presence of FBs and suggesting a repeat procedure [4]. Therefore, there is a need for a predictive model or algorithm based on clinical and radiological data to increase the index of suspicion in this category of patients. The aim of this study was to review the clinical and radiological characteristics of children under 5 years of age who underwent rigid bronchoscopy for suspected organic FBA, and to evaluate the ability of these characteristics to predict positive FBA.

Methods

Patient selection

In this study, the medical records of children who underwent rigid bronchoscopy for suspected organic radiolucent FBA between January 2018 and December 2022 were reviewed. The inclusion criteria consisted of children under 5 years of age who underwent rigid bronchoscopy to investigate suspected aspiration of organic FB. The included patients were admitted to the thoracic surgery department of a tertiary care hospital after referral from other hospitals or pediatric department. We excluded patients with definite indications for bronchotomy or flexible bronchoscopy, patients with aspirated inorganic or metallic FB, and patients on mechanical ventilation before bronchoscopy. Patients were divided into two groups based on whether or not a FB was extracted using a rigid bronchoscopy.

Diagnostic tools and measures

The suspicion of FBA was based on history of witnessed aspiration, clinical symptoms, physical signs, and radiological findings. The institutional protocol for diagnosis of children with FBA includes detailed medical history of present illness and systemic diseases in addition to systemic and local thoracic physical examination. The alarming clinical findings of FBA include symptoms of sudden cough, witnessed choking, acute respiratory distress, wheezy chest, and signs of recurrent respiratory tract infection. Initial radiological assessment is based on postero-anterior and lateral view chest X-ray.

In children with unclear history, clinical features, or radiological findings, further imaging evaluation was performed using multi-detector computer tomography (MDCT) with or without virtual bronchoscopy (VB).

Bronchoscopy procedure

Standard steps of rigid bronchoscopy were performed in the operating room under general anesthesia and controlled ventilation. The size of the bronchoscopy varied from 3 to 4.5 mm depending on the age and weight of the child. When FB was identified, forceps were inserted into the bronchoscope to grasp the FB and remove it through or with the bronchoscope, depending on the size of the FB. Follow-up after rigid bronchoscopy was performed by chest X-ray. The procedure could be repeated after 24 h if there were remnants of FB or difficulty in its removal. Also, rigid bronchoscopy could be repeated in cases with negative rigid bronchoscopy without an alternative diagnosis on imaging or clinical evaluation. In some cases where rigid bronchoscopy yielded negative results, flexible bronchoscopy was employed to evaluate more distal airways and to avoid the need for repeat rigid bronchoscopy. Most of the patients were discharged within 12 h after rigid bronchoscopy.

Data collection

The retrieved pre-procedure data include: (1) Clinical manifestations on presentation including choking, cough, diminished breath sounds, cyanosis, fever; and (2) Radiological findings on plain chest X-ray including hyperinflation, atelectasis, emphysema, consolidation, and infiltrates; and (3) Characteristics of the removed organic FB including its location and type.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistical Software version 20.0 (IBM Corp., Armonk, NY). Qualitative variables were expressed as counts and percentages and compared using the chi-square (χ2) test or Fisher's exact test. Quantitative variables were expressed as mean, standard deviation, median, and range and compared by t-test or Mann–Whitney U test. Variables associated with FB positivity in univariate analysis were further analyzed using backward logistic regression to identify significant predictors. The odds ratio (OR) and 95% confidence interval (95% CI) were calculated for each predictor. The diagnostic accuracy of pre-procedure variables was assessed by calculating area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). A P-value ≤ 0.05 was statistically significant.

Results

Among 285 children under the age of 5 years who underwent rigid bronchoscopy for FBA during the study period, a total of 228 children (80%) had suspected organic FBA. The majority of participants in this study were aged ≤ 3 years, comprising 88.15% (201 out of 228) of the total study group. The mean age was 1.98 ± 1.07 years, with a median age of 2 years and a range from 0.5 to 4.83 years. Furthermore, a significant portion of the participants were male, constituting 69.7% of the sample (n = 159). The right main bronchus was the most common location of the foreign body in 103 (51%), followed by left main bronchus in 65 (32.2%) and trachea/carina in 34 (16.8%). Extracted foreign bodies were nuts in 109 (54%), seeds in 25 (12.4%), popcorn in 18 (8.9%), bean in 15 (7.4%), lupine in 12 (5.9%), fruits in 10 (5%), meat in 8 (4%), fish bones in 3 (1.5%), and chicken bones in 2 (1%).

On presentation, children with positive FB had significantly higher frequencies of witnessed choking, wheezy chest, respiratory distress, diminished breath sounds, and hyperinflation on initial chest X-ray (Table 1). The diagnostic performance of clinical and radiological findings was estimated by sensitivity, specificity, PPV, NPV, and AUC as shown in Table 2. Age ≤ 3 years had the highest sensitivity (87%) followed by history of witnessed choking (81%) while diminished breath sounds had the highest specificity (92%). The factors with significant AUC under ROC curve (good diagnostic accuracy for FBA) were witnessed choking, wheezy chest, respiratory distress, and diminished breath sounds (Fig. 1). MDCT with or without VB was performed in 113 patients (49.5%). Chest computed tomography (CT) with VB was performed in 70 patients while chest CT without VB was performed in 43 patients. Three of the seventy patients with positive VB had negative rigid bronchoscopy (false positive) and 67 had positive rigid bronchoscopy (true positive), thus the positive predictive value of VB was 96% (67/70).

Table 1.

Initial clinical and radiological characteristics of children with positive versus negative rigid bronchoscopy for suspected foreign body aspiration

Variables	Results of rigid bronchoscopy		P-value
	Positive (n = 202)	Negative (n = 26)
Age ≤ 1.5 years	100(49.5%)	18(69.2%)	0.058
Age ≤ 3 years	177(87.6%)	24(92.3%)	0.748
Male gender	145(71.8%)	14(53.8%)	0.061
Witnessed choking	164(81.2%)	6(23.1%)	< 0.001*
Wheezy chest	119(58.9%)	5(19.2%)	< 0.001*
Cough	109(54%)	17(65.4%)	0.27
Diminished breath sounds	65(32.2%)	2(7.7%)	0.01*
Respiratory distress	92(45.5%)	3(11.5%)	0.001*
Cyanosis	11(5.4%)	3(11.5%)	0.20
Fever	8(4%)	3(11.4%)	0.11
Interval to rigid bronchoscopy > 1 day	50(24.8%)	3(11.5%)	0.13
Abnormal X ray	106(52.5%)	10(38.5%)	0.17
X-ray: Hyperinflation	66(32.7%)	3(11.5%)	0.02*
X-ray: Atelectasis	11(5.4%)	4(15.4%)	0.07
X-ray: Consolidation	12(5.9%)	3(11.5%)	0.38
X-ray: Infiltrates	13(6.4%)	4(15.4%)	0.11

^*significant difference

Table 2.

Diagnostic test analysis for detection of positive organic foreign bodies using initial clinical and radiological characteristics

Variables	Sensitivity	Specificity	PPV	NPV	AUC
Age ≤ 1.5 years	0.49	0.30	0.84	0.07	0.47
Age ≤ 3 years	0.87	0.07	0.88	0.07	0.47
Male gender	0.71	0.46	0.91	0.17	0.59
Witnessed choking	0.81	0.76	0.96	0.34	0.79
Wheezy chest	0.58	0.80	0.95	0.20	0.69
Cough	0.53	0.34	0.86	0.08	0.44
Diminished breath sounds	0.32	0.92	0.97	0.14	0.62
Respiratory distress	0.45	0.88	0.96	0.17	0.67
Cyanosis	0.05	0.88	0.78	0.10	0.47
Fever	0.04	0.88	0.72	0.10	0.46
Interval to rigid bronchoscopy > 1 day	0.24	0.88	0.94	0.13	0.56
Abnormal X ray	0.52	0.61	0.91	0.14	0.57
X-ray: Atelectasis	0.05	0.84	0.73	0.10	0.45
X-ray: Consolidation	0.05	0.88	0.80	0.10	0.47
X-ray: Hyperinflation	0.32	0.88	0.95	0.14	0.60
X-ray: Infiltrates	0.06	0.84	0.76	0.10	0.45

PPV: Positive prediction value, NPV: Negative prediction value. AUC: Area under ROC curve

Fig. 1.

Receiver operating characteristic (ROC) curves for predicting foreign body aspiration by significant clinical findings

The significant variables on univariate analysis were entered into backward stepwise logistic multivariate analysis (Table 3). The significant predictors of positive FB on multivariate analysis were witnessed chocking, wheezy chest, unilateral diminished breath sounds, and respiratory distress.

Table 3.

Multivariate logistic regression (stepwise backward method) for the predictors of positive foreign body on rigid bronchoscopy

Variables	OR	95% CI for OR		P-value
		Lower	Upper
Respiratory distress	9.50	2.25	39.98	0.002*
Wheezy chest	5.14	1.55	17.00	0.007*
Diminished breath sounds	11.55	2.16	61.75	0.004*
Witnessed choking	20.37	6.43	64.43	< 0.001*

OR: Odds ratio, CI: Confidence interval. *significant predictors

Discussion

In the present study, rigid bronchoscopy yielded positive results in 88.5% of children with suspected organic. This is consistent with the wide range of positive results reported previously, ranging from 25% to 95.8% [7–12]. The presence of respiratory distress, wheezy chest, diminished breath sounds, witnessed choking increase the chance of a FB and justifies a rigid bronchoscopy.

Signs and symptoms of FBA are often nonspecific and subtle. However, the sudden onset of choking and coughing during the acute phase of FBA is the most frequent presentation, known as penetration syndrome, which corresponds to protective respiratory reflexes in response to FBA [13]. Previously reported positivity of chocking or noticed FBA were approximately 81% for suspected tracheobronchial FBA [14], 73% for radiolucent FB [8], and 60% for organic vegetable FB [15]. Choking is often reported while the child is eating or playing.

The frequency of the most common clinical findings of FBA varies in the literature. Some authors reported decreased breath sounds as the most common manifestation in up to 57% of children with positive FBA [16], while other authors reported wheezing in 44% [4] or cough in 70% [7] of children with suspected FBA.

Suspected organic FBA was more common in children under age of 3 years and among males. These findings are consistent with other studies in literature [4, 7, 8, 15]. The right main bronchus was the most frequent location of FBs in agreement with literature as it is more straight, in line with trachea, and has short length and wide caliber [17–20]. However, some authors reported an equal distribution of aspirated FBs between the left and right main bronchus in children under 3 years of age [21–23], which could be explained by the child's position at the time of the aspiration, coughing and expelling of the FB back into the left main bronchus, or anatomical variant in the position of the carina to the left of mid-trachea. The most commonly aspirated organic FBs were nuts followed by seeds which is consistent with previous studies [8, 23, 24]. The type of aspirated organic FBs can be classified as plant or animal in origin and varies according to cultural or socioeconomic differences in world regions [19, 23].

Interestingly, lack of definitive history on presentation was reported in studies with reduced positivity of rigid bronchoscopy [10], which raises the importance of thorough history taking. History of witnessed choking had the highest sensitivity for an organic FB (81%) while diminished breath sounds on physical examination had the highest specificity (92%). Choking/aspiration is considered an important clue in predicting FBA. The literature reports a sensitivity of up to 100% and a specificity of approximately 92% for the presence of choking/aspiration [8, 25–29]. However, a recent study by Ozyuksel et al. [30] reported sensitivity of only 26.4% for witnessed choking. The reported low specificity of choking/aspiration history is associated with an increase in false-positive cases, which may be due to the expanded indication of bronchoscopy to include all FBA patients [31]. Also, our reported high specificity of diminished breath sounds (92%) is consistent with other studies [12, 32, 33].

Abnormal findings on chest X-ray had a sensitivity of 52% and a specificity of 61%. This finding is consistent with the wide reported range of 41.7% to 73% for sensitivity and 50% to 78.9% for specificity of radiological findings [12, 26, 29, 33, 34]. The positive predictive value of virtual bronchoscopy in our study was 96%. Virtual bronchoscopy is a boon for detection of FB before rigid bronchoscopy because it has a higher sensitivity than chest X-ray [35] and a high positive predictive value of up to 98% [13]. If FBA is suspected but the history is unclear and there is no radiological evidence, low-dose MDCT with or without virtual bronchoscopy may be helpful. The widespread use of CT in children with suspected FBA is controversial because of the risk of radiation exposure. However, the use of low-dose radiation and balancing the potential risks of radiation with the risks of anesthesia and airway manipulation may support this imaging modality when needed [9, 13].

In our multivariate regression analysis, significant predictors of FBA were witnessed choking, wheezy chest, respiratory distress, and decreased breath sounds. In literature, there is no unique predictor of FBA in children. Using multivariate regression analysis, Reyad et al. [19] identified relevant symptoms (breathlessness, choking, cough, or hoarseness) and signs (diminished air entry, unilateral wheezes, stridor, or crepitation) as independent predictors of FBA, thus it is less likely to have FBA in the absence of clinical symptoms. Also, Fasseeh et al. [12] reported that witnessed choking, sudden cough, new-onset or recurrent wheeze, unilateral diminished breath sounds, wheezy chest, respiratory distress, and radiographic findings of unilateral hyperinflation were the significant predictors of FBA in children. Based on these findings, history of aspiration could be used to refer the child directly to rigid bronchoscopy. However, when a witness is not available, clinical symptoms/signs and radiographic findings are important aids in deciding rigid bronchoscopy [28].

Limitations of this study include: single-center experience, relatively small sample size, and retrospective nature which is prone to selection bias and missing information. However, this study evaluates the unique predictors of organic FBA in children under age of 5 years, thus creating a reliable and simple approach for diagnosis of FBA in this age group.

In conclusion, history of choking, wheezy chest, diminished air entry, and respiratory distress were identified as significant predictors of positive organic FBA in children under 5 years of age, using logistic regression analysis.

Abbreviations

AUC

Area under curve

CI

Confidence interval

FB

Foreign body

FBA

Foreign body aspiration

MDCT

Multi-detector computer tomography

NPV

Negative prediction value

OR

Odds ratio

PPV

Positive prediction value

ROC

Receiver operating characteristic

VB

Virtual bronchoscopy

Authors' contribution

All authors whose names appear on the submission 1) made substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data; or the creation of new software used in the work; 2) drafted the work or revised it critically for important intellectual content; 3) approved the version to be published; and 4) agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

None.

Data availability

Available from corresponding author on request.

Declarations

Ethics approval

Authors' Institutional Review Board (IRB) under reference number (619/2023).

Conflict of interest

The authors declare that they have no competing interests.

Patients’ consent

Not Applicable for retrospective studies.

Statement of human and animal rights

The study has been performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.