Abstract
Background
Foreign body aspiration (FBA) is a leading cause of preventable injury and death in young children, yet diagnosis can be challenging because the choking event is not always witnessed or recognized and routine investigations have limited sensitivity. Long-duration data from Central-Eastern Europe are scarce. We characterized bronchoscopically confirmed pediatric FBA in a Polish tertiary center and factors associated with complications.
Material/Methods
In a single-center retrospective cohort, we studied children (< 18 years) with a bronchoscopically confirmed foreign body admitted over 20.5 years (2001–2021). Categorical associations were tested with the chi-square, Fisher, or Fisher–Freeman–Halton exact test, and correlations with the Spearman coefficient. A multivariable logistic regression model examined factors associated with complications, with false-discovery-rate control. A 2-sided P < 0.05 indicated significance.
Results
We analyzed data from 254 children (median age 21 months; 60.6% boys). Feeding was the commonest circumstance (57.1%), and organic objects (predominantly nuts) accounted for 73.6%. The chest radiograph was normal in 16.1% and showed the classical triad in 7.5%. Rigid bronchoscopy was definitive in 90.6%, and complications occurred in 13.0%. In multivariable analysis, older age (OR 1.88, 1.18–2.98) and a prolonged or unknown interval to bronchoscopy (OR 2.75, 1.27–5.95) were independently associated with complications, whereas foreign body type was not (OR 1.02, 0.42–2.50). Bronchiectasis and fibrotic changes occurred almost exclusively with intervals exceeding 30 days or unknown.
Conclusions
Normal initial clinical and radiographic findings did not exclude FBA in this confirmed-case cohort. Complications were independently associated with older age and a longer or unknown interval to bronchoscopy rather than with foreign body type, although causality cannot be inferred.
Keywords: Airway Obstruction; Bronchoscopy; Child, Preschool; Delayed Diagnosis; Foreign Bodies; Respiratory Tract Diseases; Retrospective Studies
Introduction
Aspiration of a foreign body into the airway is a serious pediatric emergency and an important contributor to childhood morbidity and death. When an inhaled object becomes lodged in the airway, it obstructs ventilation and gas exchange and frequently requires urgent removal [1]. The presentation is heterogeneous, but the problem is concentrated in children below 3 years of age, whose airway anatomy and exploratory behavior increase susceptibility [2]. The clinical triad of cough together with wheeze and reduced breath sounds appears in only about 1 in 6 cases, which helps explain why a prompt and confident diagnosis is so often elusive [3].
Worldwide, inhaled foreign bodies rank among the leading causes of death in children under 5 years, and organic items—particularly nuts and seeds—are the objects most often involved [4]. A 15-year United States series documented close to 306 000 pediatric choking events, of which roughly three-quarters occurred in children younger than 5 years [5]. Because of pediatric airway anatomy, aspirated objects tend to settle in the right main bronchus [6], although in the youngest children both sides are affected more evenly, reflecting the limited anatomical asymmetry at that age [7].
Diagnostic delay is clinically consequential. When retrieval is postponed, prolonged retention can drive sustained bronchial-wall inflammation and progress to structural injury, including bronchiectasis, while late diagnosis has also been linked to recurrent atelectasis and pneumonia [8,9]. Because early findings are frequently normal and a recognized or suspected aspiration is not always acted upon promptly, the interval to diagnosis and bronchoscopy is one of the few potentially modifiable determinants of outcome, and is therefore a priority for study.
Although the epidemiology of pediatric FBA has been studied extensively, long-term, single-center cohorts with uniform bronchoscopic confirmation remain limited in Central-Eastern Europe, where much of the available regional evidence dates to earlier decades, including an earlier report from our own center [10]. The present study deliberately extends that institutional experience across 2 decades. Using a 20.5-year cohort of bronchoscopically-confirmed pediatric FBA from a tertiary pulmonary referral center in Rabka-Zdrój, Poland, we aimed to characterize the clinical, radiographic, bronchoscopic, and microbiological profile of confirmed disease in this setting and to identify factors associated with complications, with particular attention to the interval to bronchoscopy. We hypothesized a priori that a longer interval to bronchoscopy—especially beyond 30 days—would be associated with more frequent structural or infectious complications.
Material and Methods
Study Design and Setting
We conducted a single-center retrospective cohort study at the National Research Institute of Tuberculosis and Lung Disorders in Rabka-Zdrój, Poland, a tertiary pediatric pulmonology referral center. We reviewed the records of children (aged < 18 years) hospitalized between December 2001 and June 2021 (20.5 years) with a foreign body confirmed within the tracheobronchial tree at bronchoscopy; foreign bodies confined to the pharynx or larynx and removed by direct laryngoscopy, without tracheobronchial bronchoscopy, were outside the scope of this study. The analytic cohort comprised 254 children (Figure 1). The study adhered to the Declaration of Helsinki and is reported following the STROBE guidelines.
Figure 1.
Participant flow and data availability. Of 254 children with a bronchoscopically confirmed tracheobronchial foreign body, the numbers with timing-, radiography-, and culture-related data available for the corresponding analyses are shown.
The study was retrospective and used data extracted from hospital medical records of children treated between December 2001 and June 2021. All bronchoscopic procedures were performed for clinical indications, and written consent for bronchoscopy and hospital treatment was obtained from the patients’ legal guardians at the time of care, according to institutional practice. The present research involved only secondary analysis of fully anonymized clinical data and did not involve any additional intervention, patient contact, or prospective recruitment. Approval for this retrospective use of anonymised data was granted by the Jan Kochanowski University Institutional Review Board (Kielce, Poland) on 15 January 2025 (approval number 12/2025).
Data Collection
Data were extracted from hospital medical records onto a standardized case-report form by a clinician familiar with the patient population, with ambiguous entries verified against the source documentation. The variables retrieved were those used in the descriptive and exploratory analyses presented here. Demographic information comprised age and sex, and aspiration-related information the circumstances of the event (eg, feeding- or play-related). For the object itself we recorded type, organic or inorganic nature, and anatomical location. Bronchoscopy data comprised the type of bronchoscope used for retrieval, any conversion from flexible to rigid bronchoscopy, completeness of removal, and the interval from suspected aspiration to bronchoscopy. Clinical history covered prior diagnoses for which the child had been treated, the referral diagnosis, and any antibiotic given within 7 days before bronchoscopy. Clinical data comprised admission symptoms and physical-examination findings, length of stay, radiographic findings, respiratory cultures obtained at bronchoscopy, in-hospital treatment, and recorded complications (pneumonia, granulation tissue, bronchiectasis, fibrotic parenchymal changes, pneumothorax, pneumomediastinum, mucosal pressure injury, airway scarring, eosinophilic bronchitis, and chronic bronchitis).
Statistical Analysis
Quantitative variables were summarized as mean (standard deviation, SD) for approximately normal distributions and median (interquartile range, IQR) otherwise, with minimum and maximum; categorical variables were summarized as counts and percentages. Unless otherwise stated, percentages were calculated for the full analytic cohort (N = 254), with “not performed” or “not obtained” retained as explicit categories (eg, chest radiography not performed in 25 children [9.8%]; respiratory cultures were not obtained in 70 [27.6%]). Several clinical variables—presenting symptoms, physical-examination findings, radiographic findings, respiratory-culture isolates, in-hospital treatments, prior and referral diagnoses, and complications—were multiple-response items, because an individual child could have more than 1 finding; for these variables the category percentages were calculated over the whole cohort and therefore sum to more than 100%. Missing data were not imputed; each analysis used all available observations for the variables involved.
Predefined age categories were ≤ 12 months (ie, up to and including 12 months of age), 13–36 months (> 1 to 3 years), 37–84 months (> 3 to 7 years), and > 84 months (> 7 years), reflecting developmental stages relevant to aspiration risk and airway anatomy; the categories were contiguous, and children aged exactly 12 months were assigned to the youngest (≤ 12-month) group, so that no child was excluded by the age classification. The interval from suspected aspiration (or onset of respiratory symptoms) to bronchoscopy was categorized as < 24 hours, 1–7 days, 8–30 days, > 30 days, and unknown; the 30-day threshold was specified a priori because prolonged retention of an airway foreign body beyond approximately 1 month has been associated with bronchial-wall injury and bronchiectasis [8,9]. For dichotomous analyses, delayed bronchoscopy was defined as a prolonged or unknown interval (> 30 days or unknown timing) versus ≤ 30 days, with a sensitivity analysis restricting delay to > 30 days (excluding unknown timing).
Categorical associations were tested with Pearson’s chi-square test (Yates’ continuity correction for 2 × 2 tables) when expected cell counts were adequate, and with exact tests otherwise: Fisher’s exact test for 2 × 2 tables and the Fisher–Freeman–Halton exact test for larger tables, computed by Monte-Carlo approximation for sparse layouts. For the principal 2 × 2 associations (bronchiectasis and absence of examination abnormality, each by foreign body type), odds ratios (OR) with 95% confidence intervals (CI) were calculated. Associations involving ordered categories or quantitative variables (eg, hospitalization duration with age) used Spearman’s rank correlation.
Because the exploratory analyses involved multiple association tests, we controlled the false discovery rate using the Benjamini–Hochberg procedure across the prespecified set of exploratory comparisons; both unadjusted and FDR-adjusted P values are reported, with FDR-adjusted significance set at q < 0.05.
To identify factors independently associated with complications, we fitted a single multivariable logistic-regression model for the composite outcome of any complication (≥ 1 recorded complication). The model was kept parsimonious in keeping with events-per-variable considerations (33 events; 3 covariates; ≈ 11 events per variable) and included age category (ordinal), foreign body type (organic vs inorganic), and delayed bronchoscopy (prolonged or unknown interval vs ≤ 30 days); adjusted ORs with 95% CIs are reported, and the > 30-day delay definition was used in a sensitivity analysis. Multivariable models for individual rare complications (eg, bronchiectasis, 6 events; fibrotic parenchymal changes, 4 events) were not fitted, because the number of events was insufficient for stable estimation and would risk overfitting and separation. Time-to-event and competing-risks analyses were not undertaken, as the retrospective records did not provide reliable longitudinal follow-up or time-to-complication data; complications were ascertained cross-sectionally during the index episode.
A two-sided P < 0.05 indicated statistical significance for individual tests. Analyses were performed in R version 4.4.1.
Results
Patient Demographics
Over 20.5 years, 254 children with bronchoscopically confirmed tracheobronchial FBA were analyzed. The median age was 21 months (IQR 16–32; range 7–181); most (65.0%) were aged 13 to 36 months (> 1 to 3 years), and the youngest children (≤ 12 months) formed the next largest group (12.6%). Boys predominated (60.6%). The age distribution is shown in Figure 2.
Figure 2.
Distribution of age at admission across predefined age categories.
The leading circumstance of aspiration was feeding (57.1%), followed by play-related choking (11.4%), simultaneous eating and play (5.1%), and other situations (9.8%); the circumstance was undocumented in 16.5%. Circumstances differed by age (P < 0.001; Table 1), with feeding-related events being most frequent in children aged 13 to 36 months (66.1%).
Table 1.
Baseline characteristics of 254 children with bronchoscopically confirmed foreign body aspiration, by age group. Values are n (%) within the age-group column unless otherwise stated. Age categories are ≤ 12 months, 13–36 months, 37–84 months, and > 84 months. P values are for the difference across age groups (Pearson chi-square, Fisher exact, or Fisher–Freeman–Halton exact test for categorical variables; Kruskal–Wallis test for length of stay). “Both” denotes 1 foreign body with organic and inorganic components.
| Characteristic | ≤ 12 mo | 13–36 mo | 37–84 mo | > 84 mo | Overall (N = 254) | P |
|---|---|---|---|---|---|---|
| n | 32 | 165 | 26 | 31 | 254 | |
| Male sex | 16 (50.0) | 100 (60.6) | 17 (65.4) | 21 (67.7) | 154 (60.6) | 0.490 |
| Foreign body type, n (%) | < 0.001 | |||||
| Organic | 21 (65.6) | 139 (84.2) | 18 (69.2) | 9 (29.0) | 187 (73.6) | |
| Inorganic | 11 (34.4) | 25 (15.2) | 8 (30.8) | 22 (71.0) | 66 (26.0) | |
| Both | 0 (0.0) | 1 (0.6) | 0 (0.0) | 0 (0.0) | 1 (0.4) | |
| Circumstance of aspiration, n (%) | < 0.001 | |||||
| Feeding | 17 (53.1) | 109 (66.1) | 12 (46.2) | 7 (22.6) | 145 (57.1) | |
| Play | 7 (21.9) | 14 (8.5) | 3 (11.5) | 5 (16.1) | 29 (11.4) | |
| Eating and play | 1 (3.1) | 9 (5.5) | 1 (3.8) | 2 (6.5) | 13 (5.1) | |
| Other | 3 (9.4) | 7 (4.2) | 3 (11.5) | 12 (38.7) | 25 (9.8) | |
| Unknown | 4 (12.5) | 26 (15.8) | 7 (26.9) | 5 (16.1) | 42 (16.5) | |
| Anatomical location, n (%) | 0.008 | |||||
| Right lung | 12 (37.5) | 91 (55.2) | 14 (53.8) | 22 (71.0) | 139 (54.7) | |
| Left lung | 12 (37.5) | 63 (38.2) | 8 (30.8) | 9 (29.0) | 92 (36.2) | |
| Central airway | 6 (18.8) | 4 (2.4) | 3 (11.5) | 0 (0.0) | 13 (5.1) | |
| Multiple sites | 2 (6.2) | 7 (4.2) | 1 (3.8) | 0 (0.0) | 10 (3.9) | |
| Interval to bronchoscopy, n (%) | 0.473 | |||||
| < 24 h | 2 (6.2) | 11 (6.7) | 2 (7.7) | 5 (16.1) | 20 (7.9) | |
| 1–7 days | 20 (62.5) | 85 (51.5) | 8 (30.8) | 13 (41.9) | 126 (49.6) | |
| 8–30 days | 2 (6.2) | 24 (14.5) | 4 (15.4) | 5 (16.1) | 35 (13.8) | |
| > 30 days | 4 (12.5) | 22 (13.3) | 5 (19.2) | 3 (9.7) | 34 (13.4) | |
| Unknown | 4 (12.5) | 23 (13.9) | 7 (26.9) | 5 (16.1) | 39 (15.4) | |
| Definitive rigid bronchoscopy | 27 (84.4) | 155 (93.9) | 23 (88.5) | 25 (80.6) | 230 (90.6) | 0.058 |
| Any complication | 2 (6.2) | 17 (10.3) | 5 (19.2) | 9 (29.0) | 33 (13.0) | 0.016 |
| Bronchiectasis | 0 (0.0) | 1 (0.6) | 1 (3.8) | 4 (12.9) | 6 (2.4) | 0.002 |
| Length of stay, days, median (IQR) | 4 (3–5) | 3 (1–−7) | 4 (1–7) | 2 (1–6) | 3 (1–6) | 0.553 |
Types and Location of Foreign Bodies
Organic objects predominated (73.6%; 187/254), inorganic objects accounted for 26.0% (66/254), and 1 child (0.4%) had both. As a proportion of the whole cohort, the commonest objects were peanuts (32.3%), other nuts (20.1%), and grain or husk fragments (4.7%); the leading inorganic items were metallic pins or needles (4.3%), foil fragments (4.3%), and toy parts (3.1%). The proportion of organic objects was highest in children aged 13 to 36 months (84.2%) and lowest above 7 years (29.0%; P < 0.001; Table 1).
By anatomical grouping, foreign bodies lay in the right-lung airways in 54.7%, the left-lung airways in 36.2%, the central airways in 5.1%, and multiple sites in 3.9%. Location varied with age (P = 0.007; Table 1): central-airway involvement was most frequent in the youngest children (≤ 12 months; 18.8%) and absent above 7 years, whereas right-lung lodgement was commonest in children older than 7 years (71.0%).
Clinical Presentation: Symptoms and Physical Examination
The leading presenting symptoms were cough (64.6%), wheezing (37.4%), and tachypnoea or respiratory distress (28.0%); fever was reported in 7.5%, and recurrent respiratory infections had preceded admission in 14.6%.
On examination, a unilateral reduction in breath sounds was the commonest finding (56.3%) and was more frequent with lateralized than with central foreign bodies (P < 0.001). Localized wheezing was present in 11.8% and diffuse wheezing in 22.4%. The full triad—cough with auscultatory wheezing and unilaterally diminished breath sounds—was documented in 19 children (7.5%). A normal physical examination was more frequent with inorganic than organic objects (7.6% vs 1.6%; odds ratio [OR] 5.03, 95% CI 1.17–21.66; P = 0.030).
Radiographic Findings
Chest radiography was not performed in 25 children (9.8%). Where obtained, the commonest findings were air trapping or hyperinflation (32.7%) and atelectasis (18.1%), with mediastinal shift in 10.2% and a radiopaque object in 8.7%. Notably, the radiograph was read as normal in 16.1% despite a confirmed foreign body. In 2 children neither the radiograph nor the examination showed any abnormality.
Bronchoscopy and Removal
Rigid bronchoscopy was performed primarily and was definitive in 71.3% of cases; flexible bronchoscopy alone sufficed in 9.4%; and in 19.3% the procedure began with flexible bronchoscopy and was converted to rigid bronchoscopy. Rigid bronchoscopy was thus the definitive technique in 90.6% of cases. Complete first-attempt removal was achieved in 73.6%, whereas 24.0% required piecemeal extraction; residual fragments necessitated a repeat bronchoscopy in 2.0%, and in 1 child (0.4%) the object could not be removed, prompting lobectomy.
Most procedures were performed 1 to 7 days after suspected aspiration (49.6%); 13.4% took place more than 30 days afterwards, and the timing was unknown in 15.4%. The distribution of the interval to bronchoscopy is shown in Figure 3.
Figure 3.
Time from suspected aspiration to bronchoscopy.
Respiratory Cultures
Respiratory cultures were not obtained in 70 children (27.6%). Normal flora was reported in 45.7%. The pathogens isolated most often were Haemophilus influenzae (12.6%), Staphylococcus aureus (5.5%), and Streptococcus pneumoniae (5.1%), with Pseudomonas aeruginosa (2.8%) and Escherichia coli (2.0%) less frequent. Culture results varied with bronchoscopy timing: P. aeruginosa was more frequent when the interval was prolonged or unknown (P = 0.012; FDR-adjusted P = 0.026), and Candida albicans was most frequent beyond 30 days (P = 0.008; FDR-adjusted P = 0.026). Culture results by timing are shown in Figure 4.
Figure 4.
Respiratory culture results. (A) Isolates across the cohort (N = 254), as the percentage of children (multiple-response: a child could yield more than 1 isolate). (B) Pseudomonas aeruginosa and Candida albicans by interval to bronchoscopy; numbers above the bars are isolate counts.
Referral Diagnoses, Prior Management, and In-Hospital Treatment
At referral, an airway foreign body was suspected in 90.9% of children; other referral diagnoses included bronchitis (5.9%), pneumonia (5.5%), and asthma (2.8%). Before the confirmed diagnosis, many children had been managed for other conditions—most often bronchitis (22.8%), pneumonia (17.3%), or asthma (7.1%); 57.9% had previously been considered to have an airway foreign body. Antibiotics had been administered within 7 days before bronchoscopy in 39.0% of children (most commonly cefuroxime or cefuroxime axetil, 13.8%, or amoxicillin–clavulanate, 10.2%), whereas 52.8% had received none and data were unavailable for 8.3%. During hospitalization, 62.2% received antibiotics, 14.2% inhaled and 12.2% received systemic glucocorticoids, and 3.5% received bronchodilators; 33.9% were managed with observation with or without respiratory physiotherapy, and 1 child (0.4%) underwent lobectomy.
Complications and Hospitalization
The median hospital stay was 3 days (IQR 1–6; mean 4.50, SD 5.1) and did not vary with age (Spearman r = −0.057, P = 0.365). At least 1 complication was recorded in 33 children (13.0%). The complications observed were pneumonia (3.5%), granulation tissue (2.8%), bronchiectasis (2.4%), fibrotic parenchymal changes (1.6%), mucosal pressure injury (1.2%), airway scarring or bridging (1.2%), pneumothorax (1.2%), pneumomediastinum (0.8%), eosinophilic bronchitis (0.4%), and chronic bronchitis (0.4%). Because some children had more than 1 complication, these categories overlap and add up to more than the overall complication rate. Freedom from complications was commonest in the youngest children (≤ 12 months) and least common above 7 years; the difference across age groups was significant (P = 0.015; Table 1).
Factors Associated with Complications
In univariable analysis, bronchiectasis was more frequent with inorganic than organic foreign bodies (4/66 [6.1%] vs 2/187 [1.1%]; OR 5.97, 95% CI 1.07–33.38; p = 0.042; FDR-adjusted P = 0.049) and differed across bronchoscopy-timing categories (P = 0.004; FDR-adjusted P = 0.026), occurring only when the interval exceeded 30 days (5.9%) or was unknown (10.3%). Fibrotic parenchymal changes followed a similar timing pattern, but the association did not reach significance on exact testing and did not survive correction for multiple comparisons (P = 0.066). Freedom from complications decreased across age groups (P = 0.015; FDR-adjusted P = 0.026).
In a multivariable logistic-regression model for any complication (33 events; 253 children with a single classifiable foreign body type), older age (adjusted OR per age-category step 1.88, 95% CI 1.18–2.98; P = 0.007) and a prolonged or unknown interval to bronchoscopy (adjusted OR 2.75, 95% CI 1.27–5.95; P = 0.010) were independently associated with complications, whereas foreign body type was not (inorganic vs organic adjusted OR 1.02, 95% CI 0.42–2.50; P = 0.96). In a sensitivity analysis restricting delay to intervals exceeding 30 days and excluding unknown timing, delayed bronchoscopy remained associated with complications (adjusted OR 3.09, 95% CI 1.20–7.96; P = 0.020). These associations are summarized in Figure 5.
Figure 5.
Forest plot of the multivariable logistic-regression model for any complication, with the sensitivity analysis for delayed bronchoscopy (adjusted odds ratios and 95% confidence intervals).
Discussion
Principal Findings
In this 20.5-year single-center cohort of 254 children with bronchoscopically confirmed tracheobronchial FBA, 3 observations stand out. First, individual findings had low sensitivity: a normal chest radiograph, a normal physical examination, and absence of the classical triad were each common among confirmed cases, so no single normal finding excluded aspiration. At the same time, the aspiration event itself was usually known—the interval to bronchoscopy could be determined in 84.6% of children, and most underwent bronchoscopy within 7 days—so the principal challenge lay less in recognizing that aspiration had occurred than in acting promptly on that suspicion, as reflected in the many children treated for other respiratory conditions before referral. Second, when complications occurred, the factors independently associated with them were older age and a prolonged or unknown interval to bronchoscopy—not the organic or inorganic nature of the foreign body. Third, selected microbiological findings (Pseudomonas aeruginosa and Candida albicans) clustered with longer or uncertain intervals, consistent with colonization of a retained foreign body. The predominance of boys and of children younger than 3 years matches other series and reflects developmental behaviors and airway anatomy that increase susceptibility [1,2,5,11–15]. These findings are descriptive and hypothesis-generating, given the retrospective, single-center, confirmed-case design.
Normal Individual Findings do Not Exclude Confirmed FBA, and Suspicion Must be Acted Upon
The low yield of individual findings in our confirmed cases reinforces a recurring clinical message: normal results do not exclude aspiration once suspicion is raised. The radiograph was read as normal in 16.1%, and the full triad was documented in only 7.5%, in keeping with the limited sensitivity of plain imaging, particularly for radiolucent organic material [3,16]. The diagnostic difficulty is also visible in the referral pathway: although an airway foreign body was suspected at referral in 90.9%—a tertiary-referral artifact—only 57.9% had previously been considered to have one, and many had been managed for bronchitis (22.8%), pneumonia (17.3%), or asthma (7.1%). Antibiotics had been administered before bronchoscopy in 39.0% of children, underscoring how often FBA is initially mistaken for infection; more recent work has proposed simplified clinical algorithms and risk-management pathways to support this judgment [17,18]. Importantly, these figures do not indicate that the aspiration itself went unrecognized: in most children the event was witnessed or otherwise known (the interval to bronchoscopy was determinable in 84.6%), and more than half underwent bronchoscopy within 7 days. The challenge is therefore less that the diagnosis is inherently obscure than that a recognized or suspected aspiration is not always acted upon promptly—consistent with the children initially treated for presumed infection before the foreign body was retrieved. Because our cohort comprises confirmed cases only, these proportions describe the spectrum of confirmed disease rather than the diagnostic accuracy of any single finding in undifferentiated suspected cases.
Delay and Age, Rather Than Foreign Body Type, Are Associated With Complications
The overall complication rate was modest (13.0%), and structural complications were uncommon (bronchiectasis 2.4%, fibrotic change 1.6%). In univariable analysis, bronchiectasis appeared more frequent with inorganic objects (OR 5.97), which might suggest a material-specific mechanism. In the multivariable model, however, the independent correlates of any complication were older age (adjusted OR 1.88 per age-category step) and a prolonged or unknown interval to bronchoscopy (adjusted OR 2.75), whereas foreign body type was not (adjusted OR 1.02). The univariable foreign body–type association is therefore most plausibly confounded by age and timing: older children and those presenting late are more likely to harbor inorganic objects and to have retained them longer. This distinction is clinically important because it moves the actionable target from the unmodifiable nature of the aspirated object toward the modifiable interval to diagnosis and retrieval. The concentration of bronchiectasis and fibrotic change in the > 30-day and unknown-timing groups is consistent with sustained bronchial-wall inflammation after prolonged retention [8,9], although the unknown-timing group probably captures the most delayed or complicated presentations, and causality cannot be inferred from this design. A practical corollary follows from the epidemiology itself. Because FBA is so strongly concentrated in children younger than 3 years, clinical awareness may be lower for older children, in whom aspiration is less expected—yet in our cohort it was precisely these older children who carried the heaviest complication burden. The very epidemiological data that this and similar series emphasize could therefore inadvertently lower suspicion in older children and contribute to the delays associated with a less-anticipated diagnosis; sustained vigilance for FBA across the pediatric age range, not only in toddlers, may help mitigate this.
The microbiological pattern fits the same framework. P. aeruginosa and C. albicans were recovered mainly when the interval was long or unknown, in line with progressive colonization of a retained foreign body and its secretions. These organisms were nonetheless infrequent (2.8% and 1.6%), and the associations rested on small numbers; they should be regarded as hypothesis-generating.
Bronchoscopic Management
Rigid bronchoscopy was the definitive technique in 90.6% of cases, consistent with ERS and ATS guidance positioning it first-line for pediatric FBA [19,20] and with its established safety in experienced hands [21,22], although it requires general anesthesia, with attendant risks that are greatest in the youngest children [23]. Flexible bronchoscopy contributed in selected situations—including the 19.3% of procedures begun flexibly and converted to rigid—reflecting its evolving role in diagnosis and in the retrieval of smaller or more distal objects [24–30]. Technique selection should remain tailored to the object, the child’s condition, and local expertise [24,26,27,29,30].
Regional and Temporal Context
The dominance of organic objects, nuts in particular, accords with contemporary European series [4,12,13]. It also marks a regional shift: older Polish data from the 1980s and 1990s were dominated by aspirated cereal grains, reflecting a more agrarian setting [10], whereas the present cohort is dominated by nuts and small snack foods. This evolution mirrors dietary and socioeconomic change and is directly relevant to locally targeted prevention.
Prevention
Our findings are compatible with caregiver-focused prevention in children under 3 years—safe feeding practices and keeping high-risk small objects out of reach, reinforced by toy-safety regulation and pediatric public-health guidance [31,32]—but, as a confirmed-case series, the study cannot quantify the effect of any preventive measure. The strongest data-driven implication concerns not a particular object but time: shortening the interval to diagnosis and bronchoscopy in children with persistent or unexplained respiratory signs is the most plausible route to fewer structural complications.
Strengths and Limitations
The principal strength of our study is a uniformly bronchoscopically confirmed cohort accrued over 2 decades at a single referral center, with a consistent diagnostic standard. Several limitations temper interpretation. The design is retrospective and single-center, and the cohort comprises confirmed cases only, without a comparison group of suspected but unconfirmed cases; the findings therefore describe confirmed disease and cannot establish the diagnostic accuracy of symptoms, signs, or radiography, nor define referral thresholds. The tertiary-referral setting inflates the proportion referred with a suspected foreign body and may under-represent milder cases. The interval to bronchoscopy was unknown in 15.4% of children; because this group carried a disproportionate share of structural complications, unknown timing likely marks the most delayed or complicated presentations and may bias timing–outcome associations. The witnessed choking event—an important diagnostic clue—was not consistently recorded and could not be analyzed, which limits inference about the value of caregiver history. Several clinical variables were multiple-response, and a number of associations rested on small cell counts with wide confidence intervals (eg, bronchiectasis by foreign body type); these should be regarded as fragile. Finally, although we fitted a multivariable model for the composite complication outcome and controlled the false discovery rate, the limited number of events precluded modeling of individual rare complications, and the findings are region-specific. Prospective, multicenter studies that include both confirmed and suspected cases and that capture the witnessed aspiration event are needed.
Conclusions
In this single-center retrospective cohort of 254 children with bronchoscopically confirmed FBA, normal initial clinical or radiographic findings occurred in a meaningful minority and therefore did not exclude an airway foreign body once it was clinically suspected. Among confirmed cases, complications were independently associated with older age and a prolonged or unknown interval to bronchoscopy, although causality cannot be inferred from this retrospective, single-center design.
Data Availability
Owing to patient confidentiality and ethical constraints, the datasets are not publicly available; they can, however, be obtained from the corresponding author on reasonable request and with IRB approval.
Footnotes
Financial support: None declared
Conflict of interest: None declared
Publisher’s note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher
Department and Institution Where Work Was Done: National Research Institute of Tuberculosis and Lung Disorders, Department of Pneumonology and Cystic Fibrosis, Rabka-Zdrój, Poland.
Ethics Approval: This study was performed in accordance with the Declaration of Helsinki. The retrospective use of fully anonymized clinical data was approved by the Institutional Review Board of Jan Kochanowski University in Kielce, Poland (approval no. 12/2025, 15 January 2025). Written consent for bronchoscopy and hospital treatment had been obtained from legal guardians at the time of care, according to institutional practice; the present study involved only secondary analysis of anonymized data, without additional intervention, patient contact, or prospective recruitment.
Declaration of Generative AI and AI-Assisted Technologies in the Manuscript Preparation Process: During the preparation of this work the authors used Claude (Anthropic) to assist with language editing and restructuring of the manuscript, with reproducible re-analysis of the authors’ own dataset (including the multivariable logistic-regression model and the false-discovery-rate correction), and with the preparation of tables and figures. All analyses were performed on data collected and owned by the authors, and the authors reviewed and verified all outputs, including every statistical result. The tool was not used to generate primary data or to create or alter source records, and is not listed as an author. The authors take full responsibility for the content of the published article.
Declaration of Figures’ Authenticity: All figures submitted have been created by the authors who confirm that the images are original with no duplication and have not been previously published in whole or in part.
References
- 1.Lowe DA, Vasquez R, Maniaci V. Foreign body aspiration in children. Clin Pediatr Emerg Med. 2015;16:140–48. [Google Scholar]
- 2.Doody D, Body F. Aspiration. In: Grillo H, editor. Surgery of the trachea and bronchi. B. C. Decker; 2004. pp. 707–18. [Google Scholar]
- 3.Midulla F, Guidi R, Barbato A, et al. Foreign body aspiration in children. Pediatr Int. 2005;47(6):663–68. doi: 10.1111/j.1442-200x.2005.02136.x. [DOI] [PubMed] [Google Scholar]
- 4.Foltran F, Ballali S, Rodriguez H, et al. Inhaled foreign bodies in children: A global perspective on their epidemiological, clinical, and preventive aspects. Pediatr Pulmonol. 2013;48(4):344–51. doi: 10.1002/ppul.22701. [DOI] [PubMed] [Google Scholar]
- 5.Chang DT, Abdo K, Bhatt JM, et al. Persistence of choking injuries in children. Int J Pediatr Otorhinolaryngol. 2021;144:110685. doi: 10.1016/j.ijporl.2021.110685. [DOI] [PubMed] [Google Scholar]
- 6.Eren Ş, Balci AE, Dikici B, et al. Foreign body aspiration in children: experience of 1160 cases. Ann Trop Paediatr. 2003;23(1):31–37. doi: 10.1179/000349803125002959. [DOI] [PubMed] [Google Scholar]
- 7.Pan H, Lu Y, Shi L, et al. Similarities and differences in aspirated tracheobronchial foreign bodies in patients under the age of 3 years. Int J Pediatr Otorhinolaryngol. 2012;76(6):911–14. doi: 10.1016/j.ijporl.2012.03.002. [DOI] [PubMed] [Google Scholar]
- 8.Karakoc F, Cakir ER, Ersu R, et al. Late diagnosis of foreign body aspiration in children with chronic respiratory symptoms. Int J Pediatr Otorhinolaryngol. 2007;71(2):241–46. doi: 10.1016/j.ijporl.2006.10.006. [DOI] [PubMed] [Google Scholar]
- 9.Mu L, He P, Sun D. The causes and complications of late diagnosis of foreign body aspiration in children: report of 210 cases. Arch Otolaryngol Head Neck Surg. 1991;117(8):876–79. doi: 10.1001/archotol.1991.01870200070010. [DOI] [PubMed] [Google Scholar]
- 10.Pogorzelski A, Zebrak J, Pawlik J, et al. Broncho-pulmonary complications due to aspiration of foreign bodies by children. Pediatr Pol. 1995;70(4):325–31. [PubMed] [Google Scholar]
- 11.Mîndru DE, Păduraru G, Rusu CD, et al. Foreign body aspiration in children—retrospective study and management novelties. Medicina. 2023;59(6):1113. doi: 10.3390/medicina59061113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Foltran F, Ballali S, Passali FM, et al. Foreign bodies in the airways: A meta-analysis of published papers. Int J Pediatr Otorhinolaryngol. 2012;76:S12–19. doi: 10.1016/j.ijporl.2012.02.004. [DOI] [PubMed] [Google Scholar]
- 13.Parvar SY, Sarasyabi MS, Moslehi MA, et al. The characteristics of foreign bodies aspirated by children across different continents: A comparative review. Pediatr Pulmonol. 2023;58(2):408–24. doi: 10.1002/ppul.26242. [DOI] [PubMed] [Google Scholar]
- 14.Pehlivan SS, Gergin ÖÖ, Sönmez E, et al. Management of foreign body aspiration in children: insights from 664 cases in the operating room: Anesthesia management in children with foreign body aspiration. The Journal of Pediatric Academy. 2023;4(3):107–12. [Google Scholar]
- 15.Tan HK, Brown K, McGill T, et al. Airway foreign bodies (FB): A 10-year review. Int J Pediatr Otorhinolaryngol. 2000;56(2):91–99. doi: 10.1016/s0165-5876(00)00391-8. [DOI] [PubMed] [Google Scholar]
- 16.Even L, Heno N, Talmon Y, et al. Diagnostic evaluation of foreign body aspiration in children: A prospective study. J Pediatr Surg. 2005;40(7):1122–27. doi: 10.1016/j.jpedsurg.2005.03.049. [DOI] [PubMed] [Google Scholar]
- 17.Fasseeh NA, Elagamy OA, Gaafar AH, et al. A new scoring system and clinical algorithm for the management of suspected foreign body aspiration in children: A retrospective cohort study. Ital J Pediatr. 2021;47(1):194. doi: 10.1186/s13052-021-01147-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Pozailov S, Goldbart A, Aviram M, et al. Foreign body aspiration score (FOBAS) – A prospectively validated algorithm for the management and prediction of foreign body aspiration in children. Eur J Pediatr. 2024;183(2):815–25. doi: 10.1007/s00431-023-05347-9. [DOI] [PubMed] [Google Scholar]
- 19.Eber E, Antón-Pacheco JL, de Blic J, et al. ERS statement: Interventional bronchoscopy in children. Eur Respir J. 2017;50(6):1700901. doi: 10.1183/13993003.00901-2017. [DOI] [PubMed] [Google Scholar]
- 20.Faro A, Wood RE, Schechter MS, et al. American Thoracic Society Ad Hoc Committee on Flexible Airway Endoscopy in Children. Official American Thoracic Society technical standards: Flexible airway endoscopy in children. Am J Respir Crit Care Med. 2015;191(9):1066–80. doi: 10.1164/rccm.201503-0474ST. [DOI] [PubMed] [Google Scholar]
- 21.Wang Q, Kong X, Wang G, et al. A real-world study of foreign body aspiration in children with 4227 cases in Western China. Sci Rep. 2024;14(1):15251. doi: 10.1038/s41598-024-65876-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.De Palma A, Brascia D, Fiorella A, et al. Endoscopic removal of tracheobronchial foreign bodies: Results on a series of 51 pediatric patients. Pediatr Surg Int. 2020;36(8):941–51. doi: 10.1007/s00383-020-04685-1. [DOI] [PubMed] [Google Scholar]
- 23.Eber E, Antón-Pacheco JL, de Blic J, et al. ERS statement: Interventional bronchoscopy in children. Eur Respir J. 2017;50(6):1700901. doi: 10.1183/13993003.00901-2017. [DOI] [PubMed] [Google Scholar]
- 24.Haller L, Barazzone-Argiroffo C, Vidal I, et al. Safely decreasing rigid bronchoscopies for foreign-body aspiration in children: An algorithm for the emergency department. Eur J Pediatr Surg. 2018;28(3):273–78. doi: 10.1055/s-0037-1603523. [DOI] [PubMed] [Google Scholar]
- 25.Tang LF, Xu YC, Wang YS, et al. Airway foreign body removal by flexible bronchoscopy: Experience with 1027 children during 2000–2008. World J Pediatr. 2009;5(3):191–95. doi: 10.1007/s12519-009-0036-z. [DOI] [PubMed] [Google Scholar]
- 26.Mansour B, Elias N. Foreign body aspiration in children with focus on the role of flexible bronchoscopy: A 5 year experience. Isr Med Assoc J. 2015;17(10):599–603. [PubMed] [Google Scholar]
- 27.Chantzaras AP, Panagiotou P, Karageorgos S, Douros K. A systematic review of using flexible bronchoscopy to remove foreign bodies from paediatric patients. Acta Paediatr. 2022;111(7):1301–12. doi: 10.1111/apa.16351. [DOI] [PubMed] [Google Scholar]
- 28.Schramm D, Ling K, Schuster A, Nicolai T. Foreign body removal in children: Recommendations versus real life – A survey of current clinical management in Germany. Pediatr Pulmonol. 2017;52(5):656–61. doi: 10.1002/ppul.23609. [DOI] [PubMed] [Google Scholar]
- 29.Safia A, Abd Elhadi U, Bader R, et al. Flexible versus rigid bronchoscopy for tracheobronchial foreign body removal in children: A comparative systematic review and meta-analysis. J Clin Med. 2024;13(18):5652. doi: 10.3390/jcm13185652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Golan-Tripto I, Weinstein Mezan D, Tsaregorodtsev S, et al. Correction to: From rigid to flexible bronchoscopy: A tertiary center experience in removal of inhaled foreign bodies in children. Eur J Pediatr. 2021;180(5):1451. doi: 10.1007/s00431-021-03960-0. [DOI] [PubMed] [Google Scholar]
- 31.EUR-Lex. Europa Publications, Eu; n.d. Directive (2009/48) http://data.europa.eu/eli/dir/2009/48/oj/eng . [Google Scholar]
- 32.Committee on Injury, Violence, and Poison Prevention. Prevention of choking among children. Pediatrics. 2010;125(3):601–7. doi: 10.1542/peds.2009-2862. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Owing to patient confidentiality and ethical constraints, the datasets are not publicly available; they can, however, be obtained from the corresponding author on reasonable request and with IRB approval.