Clinical characteristics and associated factors of post-infectious bronchiolitis obliterans in children with severe pneumonia and respiratory failure: a retrospective cohort study

Abstract

Background

Post-infectious bronchiolitis obliterans (PIBO) is a severe sequela of severe pneumonia in children. Data on its occurrence in those with concurrent respiratory failure remain limited. This study aimed to investigate the clinical characteristics and factors associated with PIBO development in this high-risk population, including an age-stratified analysis (< 2 years vs. ≥2 years).

Methods

A retrospective analysis was conducted on the clinical data of 108 children diagnosed with severe pneumonia and respiratory failure between January 2015 and June 2025. Patients were divided into a PIBO group (n = 52) and a non-PIBO group (n = 56) based on the diagnosis of PIBO at 6 months after discharge. Clinical characteristics, etiological profiles, imaging findings, and treatment modalities were compared between the two groups, and further stratified by age (< 2 years and ≥ 2 years).

Results

The prevalence of PIBO in this cohort was 48.1% (52/108). Compared to the non-PIBO group, the PIBO group had a significantly longer duration of fever during the acute phase [median 17 (IQR 12–25) days vs. 7 (3–10) days, P < 0.001], a higher prevalence of wheezing (86.5% vs. 58.9%, P = 0.001), and a significantly higher detection rate of adenovirus (ADV) (65.6% vs. 21.2%, P < 0.001). In age-stratified analysis, prolonged fever duration and invasive mechanical ventilation remained significantly associated with PIBO in both children aged < 2 years and ≥ 2 years (all P < 0.05). The PIBO group exhibited significantly higher rates of invasive mechanical ventilation (61.5% vs. 19.6%, P < 0.001) and longer ICU stays [median 14 (7–22) days vs. 6 (2–8) days, P < 0.001], and longer hospitalization [median 28 (21–51) days vs. 15 (8–21) days, P < 0.001]. No significant difference was observed between the two groups regarding a history of atopic diseases (P > 0.05). Conversely, the non-PIBO group had significantly higher incidences of liver function impairment (37.0% vs. 14.3%, P = 0.032), pulmonary consolidation on chest CT (66.1% vs. 34.6%, P = 0.001), and patchy exudation (62.5% vs. 30.8%, P = 0.001).

Conclusions

In children with severe pneumonia and respiratory failure, prolonged fever during the acute phase, ADV infection, and indicators of critical illness such as prolonged invasive ventilation and ICU support are significantly associated with the subsequent development of PIBO, and these associations were consistent across age groups. Enhanced long-term follow-up and early assessment are recommended for children presenting with these features.

Introduction

Bronchiolitis obliterans (BO) is a severe, chronic obstructive pulmonary disease secondary to lower respiratory tract injury, characterized by an inflammatory/fibrotic process affecting the small airways, leading to progressive narrowing and even obliteration of the lumen in airway generations 17–23 [1]. Although its exact incidence remains unclear, post-infectious BO (PIBO) is widely recognized as the most common cause of BO in children [2, 3]. PIBO results from severe lower respiratory tract infections and is currently associated with various pathogens, including adenovirus (ADV), influenza virus (Inf V), respiratory syncytial virus (RSV), Mycoplasma pneumoniae (MP), and measles [2, 3]. The overall prognosis of PIBO is poor, characterized by a high burden of long-term oxygen dependence and recurrent hospitalizations for repeated pulmonary infections, largely attributable to delayed diagnosis and irreversible airway fibrosis and obstruction [4]. The present study aims to analyze the clinical characteristics and identify factors associated with PIBO in children with severe pneumonia and respiratory failure, including an age-stratified comparison, thereby assisting clinicians in early recognition and intervention to improve patient prognosis.

Materials and methods

Study population

A total of 108 children aged 2 months to 156 months hospitalized in our pediatric department between January 2015 and June 2025 with a diagnosis of severe pneumonia and respiratory failure were enrolled in this study. Patients were categorized into a PIBO group and a non-PIBO group based on the development of BO at 6 months after discharge. The diagnosis of severe pneumonia was established according to the 2014 World Health Organization (WHO) guidelines [5], in conjunction with the 2024 diagnostic criteria for severe pneumonia in children issued by the Respiratory Group of the Society of Pediatrics, Chinese Medical Association. Respiratory failure was defined by arterial blood gas analysis: PaO₂ < 60 mmHg and/or PaCO₂ > 50 mmHg. The diagnosis of PIBO was made based on clinical manifestations and high-resolution computed tomography (HRCT) findings, following the diagnostic criteria proposed by Kavaliunaite et al. [6]. Of note, the HRCT scans analyzed in this study were those performed during the acute phase of severe pneumonia, rather than at the time of PIBO diagnosis, as we sought to evaluate whether early CT findings were associated with subsequent PIBO development. Malnutrition was defined as a weight-for-age Z-score < − 2.0 based on WHO child growth standards. Written informed consent was obtained from the parents or legal guardians of all participants, and the study protocol was approved by the institutional medical ethics committee (Ethics Approval No.: ES-2023-011-01).

Study design

A retrospective review of medical records was performed. Data were extracted from the electronic medical record system, including demographic information (date of birth, sex, birth history, growth and development history, past medical history, family history); duration of hospitalization; clinical features (symptoms, signs); routine laboratory findings; etiological investigations (pathogens identified by multiplex PCR from throat swabs or nasopharyngeal aspirates, culture of airway secretions); chest HRCT findings; bronchoscopy findings; respiratory support requirements and modalities; duration of respiratory support; intensive care unit (ICU) admission and length of stay; and pharmacotherapy received (including corticosteroids and intravenous immunoglobulin). All clinical symptoms (fever, cough, wheezing, and tachypnea) reported in this study were recorded within 48 h of hospital admission during the acute phase of severe pneumonia, not at the time of PIBO diagnosis. Wheezing was defined as auscultation of wheeze at any time during the acute illness. Tachypnea was defined according to age-specific respiratory rate thresholds (WHO criteria) at admission. All treatment measures, including hospitalization, ICU admission, mechanical ventilation, corticosteroids, and intravenous immunoglobulin (IVIG), were recorded from the acute phase of severe pneumonia, prior to subsequent PIBO diagnosis.

Statistical analysis

Statistical analysis was performed using SPSS version 26.0. All continuous variables were first assessed for normality using the Kolmogorov–Smirnov test. Normally distributed continuous data were presented as mean ± standard deviation (SD) and compared using the independent samples t-test. Non-normally distributed continuous data were presented as median with interquartile range (IQR) and compared using the Mann–Whitney U test. Categorical variables were expressed as frequencies (percentages) and compared using the Chi-square test for 2 × 2 contingency tables. A P-value < 0.05 was considered statistically significant.

Results

Baseline characteristics

Among the 108 children with severe pneumonia and respiratory failure, 52 (48.1%) developed PIBO during the follow-up period. The median time from the diagnosis of severe pneumonia to the diagnosis of PIBO in the PIBO group was 107 days (IQR 57–368 days). In the age-stratified comparison, 28/54 (51.9%) children aged < 2 years and 24/54 (44.4%) children aged ≥ 2 years developed PIBO, with no statistically significant difference between the two age groups (χ² = 0.335, P = 0.563). There were no statistically significant differences between PIBO and non-PIBO groups in terms of age, sex, history of prematurity, congenital heart disease, previous wheezing, personal history of atopic diseases, or family history of atopy (all P > 0.05). The prevalence of protein-energy malnutrition was significantly higher in the PIBO group than in the non-PIBO group (28.8% vs. 8.9%, P = 0.008). In the age-stratified comparison, the association between malnutrition and PIBO remained borderline significant in both the < 2-year subgroup (21.4% vs. 3.8%, P = 0.055) and the ≥ 2-year subgroup (37.5% vs. 13.3%, P = 0.081) (Table 1).

Table 1.

Baseline characteristics of the study groups

Characteristic	Non-PIBO (N = 56)	PIBO (N = 52)	t/z/χ2	P	<2 years		t/z/χ2	P	≥ 2 years		t/z/χ2	P
					Non-PIBO (N = 26)	PIBO (N = 28)			Non-PIBO (N = 30)	PIBO (N = 24)
Age (months), median (IQR)	31 (7–59)	22 (12–38)	1.704	0.092	6 (3–12)	12 (9–16)	-3.050	0.004	58 (39–72)	40 (34–49)	2.342	0.023
Male sex, n (%)	36 (64.3)	40 (76.9)	2.065	0.151	19 (73.1)	24 (85.7)	0.663	0.416	17 (56.7)	16 (66.7)	0.219	0.640
Underlying conditions, n (%)
Prematurity	11 (19.6)	8 (15.4)	0.337	0.561	3 (11.5)	3 (10.7)	0.009	0.923	8 (26.7)	5 (20.8)	0.032	0.859
Congenital heart disease	7 (12.5)	6 (11.5)	0.024	0.878	1 (3.8)	2 (7.1)	0.279	0.597	6 (20.0)	4 (16.7)	0.098	0.754
Malnutrition	5 (8.9)	15 (28.8)	7.089	0.008	1 (3.8)	6 (21.4)	3.694	0.055	4 (13.3)	9 (37.5)	3.041	0.081
History of previous wheezing, n (%)	7 (12.5)	14 (26.9)	3.581	0.058	2 (7.7)	7 (25.0)	2.908	0.088	5 (16.7)	7 (29.2)	0.591	0.442
Personal history of atopic diseases, n (%)
Eczema	10 (17.9)	10 (19.2)	0.034	0.854	6 (23.1)	5 (17.9)	0.019	0.890	4 (13.3)	5 (20.8)	0.540	0.462
Allergic rhinitis	3 (5.4)	1 (1.9)	0.189	0.664	0 (0)	0 (0)	-	-	3 (10.0)	1 (4.2)	0.662	0.416
Asthma	1 (1.8)	0 (0)	-	1.000	0 (0)	0 (0)	-	-	1 (3.3)	0 (0)	0.942	0.332
Family history of atopic diseases, n (%)	16 (28.6)	14 (26.9)	0.037	0.848	7 (26.9)	8 (28.6)	< 0.001	1.000	9 (30.0)	6 (25.0)	0.010	0.919

Clinical manifestations

All enrolled children presented with fever and cough. The clinical symptoms reported in Table 2 were documented within 48 h of hospital admission during the acute phase of severe pneumonia. Compared to the non-PIBO group, the PIBO group had a significantly longer duration of fever (17 days vs. 7 days, P < 0.001) and a higher peak body temperature (40.0 °C vs. 39.5 °C, P = 0.008). The incidence of wheezing was significantly higher in the PIBO group (86.5% vs. 58.9%, P = 0.001), whereas the incidence of tachypnea did not differ significantly between groups (94.2% vs. 94.6%, P = 1.000). In age-stratified analysis, prolonged fever duration remained strongly associated with PIBO in both the < 2-year subgroup and the ≥ 2-year subgroup. Wheezing was significantly more frequent in PIBO patients aged < 2 years but did not differ significantly in those aged ≥ 2 years. Peak temperature was significantly higher in PIBO patients aged < 2 years but not in those aged ≥ 2 years.

Table 2.

Clinical manifestations of the study groups

	Non-PIBO (N = 56)	PIBO (N = 52)	χ2/t/z	P	<2 years		χ2/t/z	P	≥ 2 years		χ2/t/z	P
					Non-PIBO (N = 26)	PIBO (N = 28)			Non-PIBO (N = 30)	PIBO (N = 24)
Symptoms
Fever duration (days), median (IQR)	7 (3–10)	17 (12–25)	-6.372	< 0.001	6 (3–8)	16 (14–19)	-5.602	< 0.001	7 (5–12)	17 (15–29)	-4.604	< 0.001
Peak temperature (°C), median (IQR)	39.5 (38.9–40.0)	40.0 (39.3–40.0)	-2.704	0.008	39.3 (38.7–39.6)	40.0 (39.2–40.0)	-3.050	0.004	40.0 (39.0-40.1)	40.0 (40.0–40.0)	-1.676	0.101
Cough, n (%)	56 (100)	52 (100)	-	-	26 (100)	28 (100)	-	-	30 (100)	24 (100)	-	-
Wheezing, n (%)	33 (58.9)	45 (86.5)	10.245	0.001	17 (65.4)	27 (96.4)	8.610	0.003	16 (53.3)	18 (75.0)	1.835	0.175
Tachypnea, n (%)	53 (94.6)	49 (94.2)	< 0.001	1.000	26 (100)	27 (96.4)	0.946	0.331	27 (90.0)	22 (91.7)	0.044	0.834
Type of infection, n (%)
Monoinfection	11 (21.2)	13 (28.9)	0.448	0.503	0 (0)	4 (14.3)	4.011	0.045	11 (36.7)	9 (37.5)	0.004	0.950
Mixed infection	41 (78.8)	32 (71.1)	0.185	0.667	23 (88.5)	18 (64.3)	3.090	0.079	18 (60.0)	14 (58.3)	< 0.001	1.000
Pathogen category, n (%)
Bacteria	36 (64.3)	24 (46.2)	3.590	0.058	23 (88.5)	13 (46.4)	8.910	0.003	13 (43.3)	11 (45.8)	< 0.001	1.000
Virus	33 (58.9)	32 (61.5)	0.077	0.782	17 (65.4)	16 (57.1)	0.117	0.733	16 (53.3)	16 (66.7)	0.507	0.476
Atypical pathogens	28 (50.0)	27 (51.9)	0.040	0.842	5 (19.2)	11 (39.3)	1.728	0.189	23 (76.7)	16 (66.7)	0.260	0.610
Fungus	11 (19.6)	14 (26.9)	0.803	0.370	7 (26.9)	8 (28.6)	< 0.001	1.000	4 (13.3)	6 (25.0)	1.203	0.273
Extrapulmonary complications, n (%)
Cardiac enzyme abnormalities	33/53 (62.3)	13/28 (46.4)	1.872	0.171	14/25 (56.0)	9/16 (56.3)	< 0.001	1.000	15/27 (55.6)	3/9 (33.3)	1.333	0.248
Liver function impairment	20/54 (37.0)	4/28 (14.3)	4.610	0.032	5/25 (20.0)	3/19 (15.8)	0.129	0.720	15/29 (51.7)	1/9 (11.1)	4.647	0.031
Renal function impairment	1/54 (1.9)	0 (0)	-	1.000	1/25 (4.0)	0 (0)	0.778	0.378	1/29 (3.4)	0 (0)	0.284	0.594
Coagulation dysfunction	29/52 (55.8)	13/27 (48.1)	0.415	0.520	9/23 (39.1)	8/18 (44.4)	0.001	0.981	21/29 (72.4)	6/9 (66.7)	0.110	0.740
Chest HRCT findings, n (%)
Pulmonary consolidation	37 (66.1)	18 (34.6)	10.675	0.001	18 (69.2)	11 (39.3)	3.732	0.053	19 (63.3)	7 (29.2)	4.941	0.026
Patchy exudation	35 (62.5)	16 (30.8)	10.893	0.001	18 (69.2)	9 (32.1)	6.008	0.014	17 (56.7)	7 (29.2)	3.046	0.081
Pleural effusion	13 (23.3)	5 (9.6)	3.590	0.058	4 (15.4)	1 (3.6)	2.239	0.135	9 (30.0)	4 (16.7)	0.670	0.413
Bronchiectasis	2 (3.6)	8 (15.4)	3.183	0.074	1 (3.8)	4 (14.3)	4.011	0.045	1 (3.3)	4 (16.7)	2.821	0.093
Pneumothorax	3 (5.4)	0 (0)	1.225	0.268	1 (3.8)	0 (0)	1.097	0.295	2 (6.7)	0 (0)	1.662	0.197
Pneumomedias-tinum	1 (1.8)	0 (0)	-	1.000	0 (0)	0 (0)	-	-	1 (3.3)	0 (0)	0.815	0.367

Regarding etiological findings, no significant differences were found between the two groups in the type of respiratory infection or the detection rates of various pathogen categories (bacteria, viruses, atypical pathogens, or fungi) in the overall comparison. All atypical pathogens identified were MP. The predominant bacteria and viruses detected differed between groups. In the non-PIBO group, the most frequently isolated bacteria were Haemophilus influenzae (30.6%), methicillin-resistant Staphylococcus aureus (22.2%), and Acinetobacter baumannii (16.7%); in the PIBO group, the most common bacteria were H. influenzae, S. pneumoniae, and A. baumannii (each 25.0%). Regarding viral pathogens, the non-PIBO group had the highest detection rates for RSV (33.3%), Inf V (24.2%), and ADV (21.2%), whereas the PIBO group had the highest rates for ADV (65.6%), cytomegalovirus (21.9%), and rhinovirus (15.6%). Of note, 12 out of 108 patients (11.1%) had no identifiable pathogen despite comprehensive testing (multiplex PCR and culture); among these, 5 (41.7%) developed PIBO and 7 (58.3%) did not. In the age-stratified analysis of etiological findings, bacteria were detected significantly more frequently in the non-PIBO group among children aged < 2 years (88.5% vs. 46.4%, P = 0.003), while no difference was observed in the ≥ 2-year subgroup.

The incidence of liver function impairment was significantly higher in the non-PIBO group overall (37.0% vs. 14.3%, P = 0.032). In the age-stratified analysis, this difference was significant only in the ≥ 2-year subgroup (51.7% vs. 11.1%, P = 0.031) and not in the < 2-year subgroup (20.0% vs. 15.8%, P = 0.720). No significant differences were observed between the groups in the incidences of cardiac enzyme abnormalities, renal impairment, or coagulation dysfunction in the overall or age-stratified comparisons (all P > 0.05).

All children underwent chest HRCT during the acute phase of severe pneumonia. The non-PIBO group had significantly higher incidences of pulmonary consolidation (66.1% vs. 34.6%, P = 0.001) and patchy exudation (62.5% vs. 30.8%, P = 0.001) compared to the PIBO group. In the age-stratified comparison, pulmonary consolidation remained significantly higher in the non-PIBO group among children aged ≥ 2 years (63.3% vs. 29.2%, P = 0.026), and patchy exudation was significantly higher in non-PIBO patients aged < 2 years (69.2% vs. 32.1%, P = 0.014). The incidence of bronchiectasis was higher in the PIBO group among children aged < 2 years (14.3% vs. 3.8%, P = 0.045), though this difference was not significant in the overall comparison (15.4% vs. 3.6%, P = 0.074). There were no significant differences between the groups in the rates of pleural effusion, pneumothorax, or pneumomediastinum in either the overall or stratified analyses (all P > 0.05) (Table 2).

Treatment measures

All treatment measures reported here were recorded during the acute phase of severe pneumonia, prior to the subsequent diagnosis of PIBO. The PIBO group had a significantly longer hospital stay (28 days vs. 15 days, P < 0.001), a higher ICU admission rate (90.4% vs. 75.0%, P = 0.036), and a longer ICU stay (14 days vs. 6 days, P < 0.001). In age-stratified analysis, hospital stay was significantly longer in the PIBO group in both the < 2-year subgroup and the ≥ 2-year subgroup. Similarly, ICU stay duration was significantly longer in the PIBO group in both age strata.

The PIBO group demonstrated a significantly higher rate of invasive mechanical ventilation (61.5% vs. 19.6%, P < 0.001), and this association was robust in both the < 2-year subgroup (53.6% vs. 15.4%, P = 0.008) and the ≥ 2-year subgroup (70.8% vs. 23.3%, P = 0.001). The duration of mechanical ventilation was also significantly longer in the PIBO group overall (14 vs. 8 days, P = 0.012), although this difference was not statistically significant in either age stratum when analyzed separately (P > 0.05 for both).

The proportion of children undergoing bronchoscopic intervention was significantly higher in the PIBO group overall (93.3% vs. 72.7%, P = 0.023), and this difference persisted in both the < 2-year subgroup (100% vs. 73.1%, P = 0.014) and the ≥ 2-year subgroup (81.8% vs. 72.4%, P = 0.035). However, no significant differences were observed between the two groups in the number of interventions or the rate of plastic bronchitis detected by bronchoscopy (both P > 0.05 in the overall comparison). In the age-stratified analysis, plastic bronchitis was detected more frequently in the non-PIBO group among children aged ≥ 2 years (23.3% vs. 4.2%, P = 0.049), while no cases were observed in either group among children aged < 2 years.

Furthermore, no statistically significant differences were found between the groups in the use of intravenous corticosteroids or IVIG, nor in either age stratum (all P > 0.05) (Table 3).

Table 3.

Treatment measures in the study groups

	Non-PIBO (N = 56)	PIBO (N = 52)	χ2/t/z	P	<2 years		χ2/t/z	P	≥ 2 years		χ2/t/z	P
					Non-PIBO (N = 26)	PIBO (N = 28)			Non-PIBO (N = 30)	PIBO (N = 24)
Length of hospital stay (days), median (IQR)	15 (8–21)	28 (21–51)	-4.456	< 0.001	15 (7–20)	26 (21–40)	-3.817	< 0.001	15 (9–25)	33 (20–60)	-2.928	0.007
ICU admission, n (%)	42 (75.0)	47 (90.4)	4.402	0.036	18 (69.2)	24 (85.7)	1.273	0.259	24 (80.0)	23 (95.8)	2.963	0.085
Length of ICU stay (days), median (IQR)	6 (2–8)	14 (7–22)	-3.752	< 0.001	7 (6–9)	15 (12–21)	-3.263	0.003	7 (4–10)	13 (8–30)	-2.588	0.016
Invasive mechanical ventilation, n (%)	11 (19.6)	32 (61.5)	19.750	< 0.001	4 (15.4)	15 (53.6)	7.027	0.008	7 (23.3)	17 (70.8)	10.336	0.001
Duration of mechanical ventilation (days), median (IQR)	8 (5–9)	14 (8–31)	-2.640	0.012	9 (8–12)	14 (8–28)	-1.137	0.271	5 (4–7)	12 (9–30)	-1.276	0.215
Bronchoscopic intervention, n (%)
Intervention	40/55 (72.7)	28/30 (93.3)	5.152	0.023	19/26 (73.1)	19/19 (100)	6.058	0.014	21/29 (72.4)	9/11 (81.8)	4.463	0.035
Number of interventions
1	25/40 (62.5)	17/28 (60.7)	0.022	0.881	13/19 (68.4)	11/19 (57.9)	0.113	0.737	12/21 (57.1)	6/9 (66.7)	0.238	0.626
≥ 2	15/40 (37.5)	11/28 (39.3)	-	-	6/19 (31.6)	8/19 (42.1)	-	-	9/21 (42.9)	3/9 (33.3)	-	-
Plastic bronchitis on bronchoscopy	7/55 (12.7)	1/30 (3.3)	1.058	0.304	0 (0)	0 (0)	-	-	7/30 (23.3)	1/24 (4.2)	3.881	0.049
Intravenous corticosteroids, n (%)	48 (85.7)	45 (86.5)	0.015	0.902	22/26 (84.6)	23/28 (82.1)	0.059	0.808	26/30 (86.7)	22/24 (91.7)	0.338	0.561
Intravenous immunoglobu—lin, n (%)	51 (91.1)	50 (96.2)	0.464	0.496	24/26 (92.3)	26/28 (92.9)	0.006	0.939	27/30 (90.0)	24/24 (100)	2.541	0.111

Discussion

PIBO is a severe chronic pulmonary sequela following severe pneumonia in children, and its development is associated with a variety of clinical factors. Previous studies have primarily focused on children infected with specific pathogens [7–9]. Children who develop severe pneumonia complicated by respiratory failure experience profound lower respiratory tract injury and physiological derangement, placing them at extremely high risk for PIBO. However, studies specifically investigating this particular high-risk subgroup remain scarce. Therefore, to the best of our knowledge, the present study is one of the first to focus specifically on this critically ill subgroup of children with severe pneumonia and respiratory failure and systematically analyze the clinical factors associated with the development of PIBO, with an additional age-stratified comparison. Although PIBO is most frequently described in children under 24 months, it can also affect older children following severe lower respiratory tract infection. Therefore, we included all pediatric patients with severe pneumonia and respiratory failure regardless of age to capture the full spectrum of this high-risk population and to allow age-stratified comparative analyses. Our age-stratified analysis did not demonstrate a statistically significant association between age < 2 years and PIBO development in this cohort. This may be explained by the uniformly critical condition of all included patients—all had severe pneumonia with respiratory failure—which may have attenuated the age-related differences typically observed in less severely ill populations. Alternatively, the dominant effect of other strong risk factors, such as ADV infection and prolonged fever, may have outweighed the influence of age.

Our findings demonstrate that children who subsequently developed PIBO had a significantly longer duration of fever, higher peak temperature, and a higher prevalence of wheezing during the acute phase of severe pneumonia. Multiple studies have confirmed that prolonged fever duration is an independent factor associated with PIBO, reflecting the detrimental effect of persistent inflammation on airway repair [9–11]. The age-stratified analysis further revealed that the association between prolonged fever and PIBO was robust across both the < 2-year and ≥ 2-year subgroups, indicating that this association is independent of age. However, the association between wheezing and PIBO was significant only in children aged < 2 years, suggesting a potential age-dependent effect. Younger children may be more susceptible to airway obstruction during severe lower respiratory infections due to their smaller airway diameter, which could amplify the clinical significance of wheezing as a marker of small airway involvement. Similarly, peak temperature was significantly higher in the PIBO group only among children aged < 2 years, which may reflect a more pronounced systemic inflammatory response to infection in younger children.

In addition, the detection rate of ADV infection was significantly higher in the PIBO group. ADV is widely recognized as the predominant pathogen responsible for PIBO in children, accounting for up to 50% of cases [2, 12]. The pathogenic mechanisms underlying ADV-associated PIBO involve direct viral injury to the airway epithelium, triggering a robust and sustained inflammatory response and aberrant repair processes [13, 14]. Importantly, even within this cohort of uniformly critically ill patients, ADV infection remained a key factor associated with subsequent PIBO. Of note, 11.1% (12/108) of patients had no identifiable pathogen despite comprehensive testing, and PIBO still developed in 41.7% of them, underscoring that the absence of an identified pathogen does not preclude the risk of PIBO and that the detection of causative agents may be limited by timing of sampling, prior antibiotic use, or low pathogen load.

Multiple indicators reflecting disease severity during the acute phase—including length of hospital stay, ICU admission rate and duration, and the rate of invasive mechanical ventilation—were all significantly higher in the PIBO group. These findings are consistent with existing evidence that hypoxemia and mechanical ventilation have been identified as strong factors associated with PIBO in several meta-analyses and clinical studies [15–17]. Importantly, all these treatment-related variables were recorded during the acute phase of severe pneumonia, before the subsequent diagnosis of PIBO, supporting their role as indicators of acute disease severity rather than consequences of an established PIBO diagnosis. The age-stratified analysis demonstrated that the associations of invasive mechanical ventilation, ICU stay duration, and hospital stay with PIBO persisted in both age strata, indicating the robustness of these associations. Of particular note, prolonged invasive mechanical ventilation is not merely a marker of critical illness; it may also directly contribute to exacerbating airway epithelial damage and subsequent fibrotic processes through ventilator-induced lung injury and extended exposure of the airways to a highly inflammatory environment [18, 19]. PICU admission has likewise been confirmed as an independent predictor of PIBO [20, 21].

Regarding baseline characteristics, the proportion of children with protein-energy malnutrition was significantly higher in the PIBO group. Malnutrition may compromise immune defense mechanisms and tissue repair capacity, thereby predisposing affected children to chronic airway sequelae following severe infection. In the age-stratified analysis, the association between malnutrition and PIBO remained suggestive in both age subgroups, although the reduced statistical power in each stratum precluded reaching conventional significance levels. This finding highlights the importance of nutritional status assessment in children with severe pneumonia. However, we did not observe a statistically significant association between a personal or family history of atopic diseases and the development of PIBO. This finding contrasts with a recent large-scale study that emphasized atopic diathesis as an independent risk factor for PIBO [22] and with other reports indicating that atopic background (including personal or family history of allergy) is significantly associated with PIBO following MP or ADV pneumonia [23, 24]. Several factors may account for this discrepancy. First, our study population was restricted to children with the most severe disease—severe pneumonia with respiratory failure—and the overwhelming severity of the acute illness may have overshadowed the influence of other potential factors. Second, the relatively small sample size and differences in the definition and assessment of atopy may also have contributed.

Another noteworthy finding was that the non-PIBO group paradoxically exhibited higher incidences of liver function impairment, pulmonary consolidation, and patchy exudation during the acute phase. In the age-stratified analysis, liver function impairment was significantly higher in the non-PIBO group only among children aged ≥ 2 years, while pulmonary consolidation and patchy exudation showed age-dependent differences as well. These findings should be interpreted with caution. The higher rate of liver function impairment in the non-PIBO group may partly reflect detection bias, as a greater proportion of non-PIBO patients underwent hepatic function testing. This discrepancy arose because many children in the PIBO group received initial treatment for severe pneumonia at other hospitals and were referred to our center only after developing PIBO, resulting in incomplete external medical records, particularly laboratory data. As such, the lower observed rate of liver function impairment in the PIBO group should be considered a potential underestimate rather than a true biological difference. Similarly, the higher rates of pulmonary consolidation and patchy exudation in the non-PIBO group, along with the higher detection rate of bacteria in this group among younger children, may reflect a greater proportion of bacterial pneumonia in the non-PIBO cohort, which typically presents with more extensive alveolar exudation and consolidation. In contrast, the pathological process predisposing to PIBO is characterized predominantly by bronchiolar epithelial injury and intraluminal fibrosis, particularly in the context of ADV infection [25]. Some studies have suggested that radiographic findings such as lobar consolidation in severe ADV pneumonia are also associated with an increased risk of PIBO [26].

The high rates of systemic corticosteroid and IVIG use (approximately 80–90% in both groups) reflect the uniformly critical condition of this cohort, in which all patients had severe pneumonia complicated by respiratory failure. At our institution, these therapies are administered as adjunctive immunomodulatory treatment for severe sepsis or suspected hyperinflammatory states during the acute phase, and were not intended as PIBO prophylaxis. Importantly, despite these high utilization rates, neither therapy showed a significant association with PIBO development in this study, possibly due to the near-universal use in both groups limiting the statistical power to detect a difference. The therapeutic efficacy of these interventions in preventing PIBO warrants further investigation in prospective studies.

In summary, this study examined the clinical factors associated with the development of PIBO in the extremely high-risk population of children with severe pneumonia and respiratory failure. Our results demonstrate that prolonged fever during the acute phase, ADV infection, and multiple indicators of critical illness are significantly associated with subsequent PIBO, and many of these associations remained robust across age strata. These findings suggest that children presenting with these features should be regarded as being at very high risk for PIBO. While actively managing the acute illness, clinicians should establish systematic long-term follow-up plans, closely monitor respiratory symptoms, and promptly perform pulmonary function tests and chest HRCT assessments to facilitate early identification and intervention for PIBO, thereby improving long-term outcomes. As a single-center retrospective study, our findings require validation through multicenter, prospective investigations.

Abbreviations

ADV

adenovirus

BO

Bronchiolitis obliterans

HRCT

high-resolution computed tomography

ICU

intensive care unit

Inf V

influenza virus

IQR

interquartile range

MP

Mycoplasma pneumoniae

PIBO

post-infectious BO

RSV

respiratory syncytial virus

SD

standard deviation

WHO

World Health Organization

Authors’ contributions

XC performed collection of data, statistical analysis, drafted the article, and edited the manuscript. HL, SW and KZ performed data collection and revised the manuscript. ST, JX, YP and YW performed the data statistics and analysis. DC conceptualized and designed the study, performed data analysis, and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Major Clinical Research Project of the Scientific Research Capacity Enhancement Program of Guangzhou Medical University [Grant number GMUCR2024-02008] and the General Guidance Project of Guangzhou Health Science and Technology [Grant number 20251A011067].

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou Medical University (reference number: ES-2023-011-01). Written informed consent to be included in the study was obtained from the parents of the included children.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.