Transcatheter embolization of systemic artery-to-pulmonary artery shunts with n-butyl-2-cyanoacrylate for pediatric hemoptysis: a single-center retrospective study

Abstract

Background

Systemic artery-to-pulmonary artery (SA-PA) shunt represents a critical pathological mechanism underlying hemoptysis in pediatric patients, yet focused interventional research targeting this specific condition remains limited. Traditional embolic materials are frequently associated with elevated recurrence rates in SA-PA shunt cases. This study assesses the effectiveness and safety profile of n-butyl-2-cyanoacrylate (NBCA) embolization for hemoptysis secondary to pediatric SA-PA shunts, while examining the clinical significance of individualized NBCA dilution protocols.

Methods

This study retrospectively reviewed pediatric hemoptysis cases treated with NBCA embolization for SA-PA shunts at a single center from May 2021 to March 2025. Baseline characteristics, culprit feeding artery features, and NBCA-Lipiodol mixing ratios were collected. The primary study endpoints were technical success rate, 24-hour clinical success rate, incidence of severe complications, and hemoptysis recurrence rate. The Kaplan-Meier analysis was applied to generate the hemoptysis-free survival curve.

Results

The study cohort comprised 26 pediatric patients (mean age, 9.12 ± 3.44 years; 12 males and 14 females). Fifty-one abnormal systemic feeding arteries were embolized: 16 (31.4%) orthotopic bronchial arteries, 9 (17.6%) ectopic bronchial arteries, and 26 (51.0%) non-bronchial systemic arteries. The mixing ratio of NBCA-Lipiodol varied between 1:2 and 1:4. The rates of technical success and 24-hour clinical success both reached 100.0%. The early (≤ 1 year) hemoptysis recurrence rate was 7.7%, with no late hemoptysis recurrence over a median 25-month follow-up. Kaplan-Meier analysis showed that median hemoptysis-free survival was not reached.

Conclusions

NBCA embolization achieved favorable safety and efficacy profiles in managing SA-PA shunt-related hemoptysis among pediatric patients, demonstrating sustained clinical benefit with limited disease recurrence. It is a valuable interventional treatment option worthy of clinical promotion.

Background

Pediatric hemoptysis is a critical emergency in pediatric clinical practice, which can be life-threatening due to airway obstruction or massive blood loss in severe cases [1]. Transcatheter embolization is now established as the preferred initial intervention for managing hemoptysis in adults [2, 3]. However, the etiology and vascular anatomical characteristics of hemoptysis in children differ significantly from those in adults, and relevant interventional treatment studies remain limited. The lack of unified standards in clinical practice poses considerable challenges to the standardized interventional management of pediatric hemoptysis [4, 5].

Systemic artery-to-pulmonary artery (SA-PA) shunt is one of the important pathological features of pediatric hemoptysis [6]. This shunt is a pathological abnormal communication between the systemic circulation and the pulmonary circulation, which can originate from congenital SA-PA fistulas or be secondary to underlying lesions such as pulmonary infections and congenital heart diseases [6–8]. Due to the significantly higher systemic systolic pressure than pulmonary pressure, the shunt presents unidirectional perfusion from the systemic circulation to the pulmonary circulation. Vessels in the shunt area are subjected to chronic high pressure, making them highly susceptible to rupture and bleeding, which is also a risk factor for recurrent hemoptysis. Therefore, the core principle of interventional treatment for SA-PA shunt is to achieve complete and permanent embolization of the culprit feeding arteries and shunt orifices.

Gelatin sponge, polyvinyl alcohol (PVA) particles, and coils represent the predominant embolic agents in current clinical practice [9–12]. However, a recent scoping review by Zietarska et al. demonstrated that these agents are associated with relatively high long-term recurrence rates in the management of hemoptysis [13]. Especially in cases with SA-PA shunts, particulate embolic agents are easily carried into the pulmonary artery by high-velocity blood flow and cannot effectively remain at the shunt orifice [14]. Coils can only achieve embolization of the main trunk of the culprit vessel [15, 16], which failing to completely occlude the shunt orifice and distal abnormal vasculature, resulting in high risks of postprocedural vascular recanalization and neovascularization, with limited long-term effectiveness. As a liquid embolic material, n-butyl-2-cyanoacrylate (NBCA) can polymerize rapidly after contacting blood to form a permanent vascular cast [17], which is theoretically more suitable for embolization of tortuous blood vessels and high-flow shunt lesions, and is expected to reduce the risk of postprocedural hemoptysis recurrence from the pathological mechanism.

NBCA embolization has emerged as a reliable technique for adult hemoptysis during the last several years [18–20]. However, in current pediatric-related studies, there is neither a systematic analysis of NBCA embolization specifically for children with SA-PA shunt, nor a clear individualized mixing ratio strategy and safe operation boundary for this material in pediatric SA-PA shunt embolization. Meanwhile, long-term prognostic data for this high-recurrence-risk pathological mechanism in the pediatric population are extremely scarce. Therefore, this single-center retrospective analysis was conducted to investigate the clinical effectiveness and safety of NBCA embolization for SA-PA shunt-related hemoptysis in children, providing clinical evidence for the standardized interventional diagnosis and treatment of these patients.

Methods

Study design and participants

Pediatric patients presenting with hemoptysis and undergoing transcatheter NBCA embolization for SA-PA shunts were identified at our institution between May 2021 and March 2025 in this retrospective single-center observational analysis. Inclusion criteria: (1) Age ≤ 18 years; (2) Digital subtraction angiography (DSA) confirmed the presence of at least one SA-PA shunt (defined as early opacification of pulmonary artery branches), which was the culprit lesion for hemoptysis; (3) Culprit feeding arteries included orthotopic bronchial arteries, ectopic bronchial arteries, and non-bronchial systemic arteries; (4) Only a mixture of NBCA and Lipiodol was used as the embolic material during the operation; (5) All patients had a postprocedural follow-up period of at least 12 months, with complete clinical and imaging data regardless of whether hemoptysis recurrence occurred. Exclusion criteria: (1) Patients who underwent surgical pulmonary resection or vascular ligation due to hemoptysis-related causes within 12 months after embolization, making it impossible to evaluate the long-term effectiveness of embolization; (2) Loss to follow-up before 12 months postprocedure, regardless of whether hemoptysis recurrence occurred; (3) Severe lack of clinical or imaging data; (4) Patients who concurrently used other embolic materials such as coils, gelatin sponge, or PVA during the embolization; (5) Patients with systemic artery-to-pulmonary venous shunts (defined as early opacification of a pulmonary vein or the left atrium on DSA).

Twenty-six children fulfilling the predefined selection criteria constituted the final study cohort. Figure 1 presents representative computed tomography and DSA findings of a pediatric patient with hemoptysis. Figure 2 illustrates different angiographic patterns of SA-PA shunts.

Fig. 1.

Representative computed tomography and digital subtraction angiography (DSA) findings in a child with hemoptysis. a Axial lung window computed tomography shows extensive patchy hyperdense lesions in the right upper lobe (black arrow). b Coronal maximum intensity projection demonstrates the right bronchial artery coursing toward the lesion site (white arrow). c Volume rendering demonstrates the spatial location and course of the right bronchial artery (white arrow). d DSA clearly depicts the enlarged bronchial artery (short white arrow), the location of the systemic artery-to-pulmonary artery (SA-PA) shunt orifice (long white arrow), and early opacification of the right pulmonary artery branches (black arrows)

Fig. 2.

Different angiographic patterns of SA-PA shunts. a Ectopic bronchial artery arising from the right internal thoracic artery: the arterial trunk is enlarged and tortuous (white arrow), with early opacification of the right lower pulmonary artery (black arrows). b Right orthotopic bronchial artery with a tortuous course (white arrow) and focal dilatation, accompanied by early opacification of the right lower pulmonary artery (black arrow). c Non-bronchial systemic artery arising from the right internal thoracic artery (white arrow), with early opacification of the right upper pulmonary artery (black arrows). d Non-bronchial systemic artery arising from the right inferior phrenic artery (white arrow), showing early opacification of the right pulmonary artery branches (black arrows)

Data collection

Core data were retrospectively retrieved from our hospital’s picture archiving and communication and electronic medical record systems, as well as follow-up records: (1) Baseline patient data: demographic data, preprocedural hemoptysis volume grade (using pediatric weight-corrected standard grading [21]: mild hemoptysis was defined as 24-hour blood loss < 0.71 mL/kg, moderate as 0.71–3.56 mL/kg, and massive as > 3.56 mL/kg), and anatomical location of the lesion; (2) Characteristics of culprit feeding arteries: anatomical type, number, and course; (3) Procedural parameters: anesthesia method, puncture approach, and NBCA-Lipiodol mixing ratio; (4) Effectiveness, safety, and follow-up outcome data: technical and 24-hour clinical success rates, procedure-related complications, follow-up period, hemoptysis recurrence (including recurrence time and cause), and survival outcomes.

Interventional procedure

Two interventional radiologists, X.L. (11 years of experience) and Z.S. (6 years of experience), jointly conducted all procedures. Children who could not cooperate with the procedure underwent general anesthesia with combined intravenous and inhalation anesthesia, while those who could cooperate received local infiltration anesthesia with 2% lidocaine at the puncture site.

Following successful femoral arterial access with the Seldinger technique, a 5 F vascular sheath (Terumo, Tokyo, Japan) was positioned. A 5 F pigtail catheter (Cordis, Florida, USA) was used to perform aortography of the aortic arch, thoracic aorta (at the level of the T5 vertebra), and abdominal aorta (at the level of the T11 vertebra) to comprehensively screen for culprit feeding arteries and avoid missing non-bronchial systemic arteries. After identifying the culprit feeding arteries, a 5 F C2 catheter (Cordis, Florida, USA) was used for selective catheterization of the culprit feeding arteries. Angiography was conducted to clarify the course of the culprit artery, shunt volume, pulmonary artery opacification, and shunt orifice location. Subsequently, a 1.7 F Echelon 10 microcatheter (EV3, Irvine, California, USA) was introduced coaxially through the diagnostic catheter, and the microcatheter was advanced optimally near the shunt orifice, avoiding normal branch vessels such as spinal arteries and intercostal arteries.

Based on the blood flow velocity and shunt volume of the culprit feeding artery indicated by angiography, the NBCA (Glubran2, GEM, Viareggio, Italy) and Lipiodol (Guerbet, Aulnay-sous-Bois, France) mixture was individually prepared with mixing ratios ranging from 1:2 to 1:4. Mixing ratio principle: 1:2 ratio for high-flow shunts near the fistula orifice, 1:3 ratio for medium-flow vessels, and 1:4 ratio for distal vessels with tortuous courses. The microcatheter lumen was pre-flushed multiple times using 5% glucose before the mixture was delivered at a constant, gradual rate with roadmap guidance. Injection was immediately stopped when the mixture reached 1 cm proximal to the microcatheter tip or slight reflux occurred. The microcatheter was then rapidly withdrawn to prevent adhesion. After embolization, angiography of the culprit feeding artery was repeated to evaluate shunt occlusion. All procedures were performed under a pediatric low-dose protocol with pulsed fluoroscopy, strict collimation, and minimization of magnification views to reduce radiation exposure. Figure 3 demonstrates the NBCA embolization procedure in a representative pediatric patient.

Fig. 3.

NBCA embolization procedure in a representative child. a DSA demonstrates a tortuous right bronchial artery (white arrow), with early opacification of the right inferior pulmonary artery branches (black arrows), indicating an SA-PA shunt. b The microcatheter is superselectively advanced to the proximal end of the shunt orifice (white arrow). c A total of 0.8 mL of the mixture of NBCA and Lipiodol (volume ratio 1:3) is used to embolize the shunt orifice and abnormal vascular bed; fluoroscopy shows homogeneous dispersion of the embolic agent within the culprit vessels (black arrow). d Post-embolization angiography shows complete occlusion of the shunt, with no opacification of the distal bronchial artery, shunt orifice, and corresponding pulmonary artery branch

Endpoint definitions

To assess the effectiveness and safety of NBCA embolization, the primary endpoints were defined as follows: (1) Technical success: Super-selective embolization of all angiographically confirmed culprit feeding arteries, with complete occlusion of the SA-PA shunt and no early pulmonary artery opacification on postprocedural angiography; (2) Clinical success: No new active hemoptysis occurred within 24 h after embolization, and the original hemoptysis symptoms were completely relieved; (3) Severe complications: Utilizing the Society of Interventional Radiology (SIR) criteria for adverse event classification [22], any adverse event of grade C or above, including spinal or intracranial ectopic embolism, major bleeding requiring surgical treatment/blood transfusion, symptomatic pulmonary embolism, death, etc.; (4) Hemoptysis recurrence: Reappearance of active hemoptysis related to the target lesion after embolization. Recurrence was stratified by the 1-year postprocedural threshold, categorized as early if hemoptysis appeared within this period, and late if observed thereafter. This 1-year threshold is widely adopted in pediatric hemoptysis intervention studies, as the vast majority of postprocedure recurrences occur within 12 months. (5) Hemoptysis-free survival: Time from the day of embolization completion to the first occurrence of hemoptysis recurrence or all-cause death. Patients without hemoptysis recurrence or death were censored at the last follow-up time.

Follow-up protocol

Standardized follow-up was completed by all patients after the procedure, including outpatient visits, chest imaging evaluation, and telephone follow-ups. Scheduled follow-up evaluations were conducted at 1, 3, 6, and 12 months postprocedure, followed by annual assessments. The final date for follow-up data acquisition was March 2026, at which time an additional assessment was performed for all patients to capture their status up to this date. All included patients completed at least 12 months of standardized postprocedural follow-up, with no loss to follow-up and complete primary endpoint data during the follow-up period.

Statistical analysis

Statistical analyses were conducted utilizing SPSS 26.0 (IBM, Armonk, NY, USA). Categorical variables are presented as frequencies and percentages [n (%)]. For continuous variables, data normality was evaluated via the Shapiro-Wilk test. Normally distributed data are reported as mean ± standard deviation (SD), whereas non-normally distributed data are expressed as median with interquartile range (IQR). The Kaplan-Meier method was employed to estimate postprocedural hemoptysis-free survival rates in the study cohort. Patients who remained free of hemoptysis recurrence or all-cause death at their final follow-up assessment were treated as censored observations. Given the single-arm observational design of this study, no formal between-group statistical comparisons were undertaken. All analyses were based on complete cases without any imputation for missing data. A two-sided P-value < 0.05 was considered statistically significant.

Results

Baseline characteristics of the study population

This study enrolled a cohort of 26 pediatric patients presenting with hemoptysis, comprising 12 males and 14 females. The mean age of the participants was 9.12 ± 3.44 years (range: 2–16 years), with data normality verified by the Shapiro-Wilk test. The etiologies were classified into congenital SA-PA fistulas and secondary SA-PA shunt. The latter was primarily associated with pulmonary inflammation, congenital heart disease, and bronchiectasis. Preprocedural hemoptysis volume was mainly mild and moderate. Lesions were more common in the right lung. Most patients with hemoptysis had 2 or 3 culprit feeding arteries. The complete baseline characteristics of the study population are shown in Table 1.

Table 1.

Baseline Characteristics of Included Pediatric Patients

Characteristics	Total (N = 26)
Sex, n (%)
Male	12 (46.2%)
Female	14 (53.8%)
Age, years, mean ± SD (range)	9.12 ± 3.44 (2–16)
Underlying etiologies of hemoptysis, n (%)
Pulmonary inflammation	12 (46.2%)
Congenital heart disease	7 (26.9%)
Systemic artery-pulmonary artery fistula	4 (15.4%)
Bronchiectasis	3 (11.5%)
Hemoptysis grade*, n (%)
Mild hemoptysis	12 (46.2%)
Moderate hemoptysis	10 (38.5%)
Massive hemoptysis	4 (15.4%)
Number of culprit feeding arteries, n (%)
1	7 (26.9%)
2	13 (50.0%)
3	6 (23.1%)
Lesion location, n (%)
Right lung	15 (57.7%)
Left lung	11 (42.3%)

*Percentages may not sum exactly to 100% due to rounding. Mild hemoptysis was defined as blood loss < 0.71 mL/(kg·d), moderate hemoptysis as 0.71–3.56 mL/(kg·d), and massive hemoptysis as > 3.56 mL/(kg·d). SD = standard deviation

Embolization parameters

A total of 51 culprit feeding arteries in 26 children underwent NBCA embolization. All culprit feeding arteries showed enlargement, tortuosity, and SA-PA shunting. Non-bronchial systemic arteries accounted for the highest proportion. The NBCA-Lipiodol mixing ratio ranged from 1:2 to 1:4, and the dilution ratio was individually selected based on vessel course and blood flow characteristics. The distribution of culprit feeding arteries and embolization parameters are summarized in Table 2.

Table 2.

Culprit Feeding Artery Types and Embolization Parameters

Procedural Parameters	Total (N = 51)*
Anatomical types of culprit feeding arteries, n (%)
Orthotopic bronchial arteries	16 (31.4%)
Ectopic bronchial arteries	9 (17.6%)
Non-bronchial systemic arteries	26 (51.0%)
NBCA-Lipiodol mixing ratio, n (%)
1:2 dilution	13 (25.5%)
1:3 dilution	20 (39.2%)
1:4 dilution	18 (35.3%)

*N = 51 refers to the total number of embolized culprit feeding arteries in 26 patients

Effectiveness and safety outcomes

All culprit feeding arteries were effectively embolized. Postprocedural angiography showed disappearance of the SA-PA shunts and no early opacification of the pulmonary artery, with a technical success rate of 100.0%. All patients had no new active hemoptysis within 24 h after embolization, and pre-existing hemoptysis symptoms were completely relieved, with a clinical success rate of 100.0%. Two patients had a small amount of blood-streaked sputum within 24 h after embolization, which was due to the discharge of old blood clots in the airway without fresh active bleeding. No special treatment was given, and no related symptoms occurred again before discharge or during follow-up, which met the clinical success criteria.

During the follow-up period, there were 2 cases of early recurrence, corresponding to a hemoptysis recurrence rate of 7.7%. The two early recurrences were distributed across two different etiologies: one occurred in a patient with congenital heart disease, and one in a patient with bronchiectasis. No early recurrence was observed in patients with pulmonary inflammation or systemic artery-pulmonary artery fistula. No late recurrences were observed in all patients. Notably, the two recurrent cases arose from distinct mechanisms. One patient with congenital heart disease experienced recurrent mild hemoptysis on day 40 after embolization. During the second interventional procedure, it was found that an non-bronchial systemic artery originating from the inferior phrenic artery that was slightly opacified during the first procedure was missed for embolization. Another patient with bronchiectasis experienced recurrent moderate hemoptysis 7 months post-embolization. Repeat angiography showed the formation of new collateral vessels originating from the intercostal artery accompanied by early pulmonary artery opacification. NBCA embolization was performed after super-selective catheterization of the new collateral vessels with a microcatheter. No further hemoptysis recurrence occurred in the above 2 patients during the follow-up period after the second embolization.

No severe procedure-related complications occurred in the entire cohort. No ectopic embolization of central nervous vessels such as vertebral arteries, internal carotid arteries, or spinal arteries was observed intraprocedurally. In one case, the culprit feeding artery was extremely tortuous. During microcatheter superselection, vascular rupture with contrast extravasation occurred. The microcatheter was kept in place, and a mixture of NBCA and Lipiodol (1:4) was injected through it to seal the rupture. Repeat angiography after embolization showed no further contrast extravasation. No subsequent hemorrhagic complications occurred, and the final technical success of the procedure was not compromised. This event did not meet the criteria for SIR adverse event classification Grade C and was therefore excluded from the statistics of severe complications. The procedural effectiveness and safety outcomes are summarized in Table 3.

Table 3.

Effectiveness and Safety Outcomes

Outcome Measures	Total (N = 26)
Technical success, n (%)	26 (100.0%)
24-hour clinical hemostasis, n (%)	26 (100.0%)
Early recurrence (≤ 1 year postprocedure), n (%)	2 (7.7%)
Late recurrence (> 1 year postprocedure), n (%)	0 (0%)
Severe procedure-related complications, n (%)	0 (0%)

The total follow-up time of the 26 children ranged from 12.0 to 58.0 months. Since it did not conform to normal distribution, it was described as a median follow-up time of 25.0 months (IQR: 18.0–33.0 months). No deaths occurred during the follow-up period. During the follow-up period, only 2 hemoptysis recurrence events (7.7%) were observed. Consequently, the median hemoptysis-free survival time was not reached, as the cumulative event rate remained below 50%. The Kaplan-Meier hemoptysis-free survival curve is shown in Fig. 4.

Fig. 4.

Kaplan-Meier curve of hemoptysis-free survival in 26 pediatric patients treated with NBCA embolization for SA-PA shunts. The median hemoptysis-free survival time was not reached over a median follow-up period of 25 months

Discussion

This study validates NBCA embolization as a safe and effective intervention for pediatric hemoptysis driven by SA-PA shunts. Given the pathological nature of these shunts, our findings highlight the importance of targeted embolization in managing this high-hemoptysis recurrence-risk subtype.

The selection of embolic materials is an important factor affecting the hemoptysis recurrence rate after interventional embolization for hemoptysis [13, 23]. Zhang et al. have confirmed in 420 adult patients with hemoptysis that patients with SA-PA shunts have a 2.782-fold higher risk of hemoptysis recurrence after embolization than those without shunts [14]. After traditional PVA particle embolization, particles are susceptible to displacement due to bidirectional blood flow impact, leading to target vessel recanalization and long-term hemoptysis recurrence. Ryuge et al. demonstrated that recanalization (45.2%) was the predominant mechanism of recurrent hemoptysis after coil embolization, and larger-diameter arteries were significantly more prone to recanalization than smaller ones [12]. Meanwhile, occlusion of the main trunk of the culprit vessel will also cause operational difficulties for secondary interventional treatment after hemoptysis recurrence [2, 24].

In this study, the application of NBCA achieved 100% technical and 24-hour clinical success rates. No severe procedure-related complications occurred across the entire cohort. The early recurrence rate was only 7.7%, and no late recurrence related to vascular recanalization was observed. These findings fully highlight the unique advantages of NBCA in the embolization of SA-PA shunt lesions. First, NBCA rapidly polymerizes upon contact with blood to form a stable vascular cast, achieving permanent occlusion of the shunt and abnormal vascular bed [20]. This mechanism may reduce the risk of recanalization from a pathological standpoint, which likely contributes to the absence of late recurrence within the follow-up period of our study. Second, NBCA embolization has excellent operational flexibility and anatomical adaptability. Lipiodol serves as a radiopaque contrast agent and a polymerization modulator, allowing for dose-dependent control over the NBCA polymerization rate. This offers tailored maneuverability for treating lesions across diverse hemodynamic and anatomical scenarios. In this study, for culprit feeding arteries with extremely tortuous courses where the microcatheter could not be super-selected to the proximal end of the shunt orifice, a 1:4 high dilution ratio (low NBCA concentration) mixture was used. By prolonging the NBCA polymerization time, it could still disperse to the distal vascular bed under blood flow impact to achieve effective embolization. This is consistent with the conclusion of Kolu et al. that using a more diluted NBCA-Lipiodol mixture can significantly improve the success rate of embolization in tortuous vessels [25]. For culprit feeding arteries with visible fistula orifices where the microcatheter tip could approach the fistula orifice, a 1:2 low dilution ratio (high NBCA concentration) was used to achieve rapid and precise embolization of the fistula orifice, avoiding NBCA being flushed into the pulmonary artery by high-velocity blood flow. In addition, NBCA can rapidly achieve rupture site occlusion in emergency situations of intraprocedural vascular rupture, also reflecting its clinical value in acute bleeding scenarios.

Current interventional treatment studies for pediatric hemoptysis are mostly small-sample case series, and no separate analysis has been conducted for the high-recurrence-risk subtype of SA-PA shunt. This study fills an important gap in the existing literature. Kuang et al. performed bronchial artery embolization using vascular plugs in 20 pediatric patients with hemoptysis [21]. Although immediate hemostasis was achieved, the 1-year recurrence rate reached as high as 40%, which is higher than the 7.7% early recurrence rate observed in our study. Considering the distinct embolic materials and possible heterogeneity of patient cohorts between studies, this comparison requires careful interpretation. Notably, the 100% technical success rate refers to the complete embolization of all angiographically confirmed culprit feeding arteries during the initial procedure. The single case of recurrence due to a missed slightly opacified non-bronchial systemic artery highlights the importance of comprehensive vascular screening rather than indicating a failure of the embolization technique itself. By contrast, the other case of recurrence arising from a new collateral vessel represents disease-related vascular changes, rather than a shortcoming of intraprocedural angiographic screening. Meanwhile, in our study, non-bronchial systemic arteries accounted for 51.0% of all culprit feeding arteries, which was significantly higher than the 20.0% reported in the adult study [17], suggesting that the distribution of culprit feeding arteries for hemoptysis has clear population-specific characteristics. It has been reported in the literature that missing non-bronchial systemic arteries is one of the main factors for recurrence after bronchial artery embolization [26]. One case of early recurrence in this study was due to missing a slightly opacified non-bronchial systemic artery during the first embolization procedure, directly indicating the importance of comprehensive intraprocedural vascular screening for interventional treatment of pediatric hemoptysis.

The safety concerns of clinical application of NBCA are non-target tissue necrosis and microcatheter adhesion [27–29]. However, no related adverse events occurred in the entire cohort of this study, consistent with the safety results of recent relevant clinical studies [17, 18, 30]. This study effectively avoided the above risks through precise preprocedural vascular evaluation, intraprocedural super-selective microcatheter intubation, individualized NBCA-Lipiodol mixing ratio based on blood flow characteristics, and a standardized operation process of flushing the catheter with 5% glucose solution, confirming the safety and controllability of this technique in the pediatric population.

Several limitations should be acknowledged. First and foremost, the single-institution, retrospective design combined with a limited cohort size introduces potential selection bias, which may constrain the external validity of our findings. Second, it is a single-group case series without a concurrent control group of other embolic materials such as PVA and coils; therefore, a direct comparison of the effectiveness and safety of different materials was not performed. Third, no subgroup analysis was conducted for different underlying primary diseases, and the vascular characteristics and recurrence risk of shunt lesions caused by different etiologies may differ. Additionally, recurrence was determined solely by clinical hemoptysis symptoms, without routine postprocedural angiography or computed tomography angiography. Considering radiation exposure and poor compliance in children, such imaging was not performed, so subclinical recurrence and vascular recanalization may be missed. Furthermore, detailed radiation dose metrics were not available for this retrospective cohort. In the future, multicenter, large-sample prospective controlled studies are still needed to further verify the long-term effectiveness and safety of NBCA embolization of SA-PA shunts for pediatric hemoptysis and promote the establishment of a standardized system for interventional treatment of pediatric hemoptysis.

Conclusions

In conclusion, this study confirmed that NBCA embolization of SA-PA shunt for the treatment of pediatric hemoptysis can achieve a high immediate hemostasis success rate and low hemoptysis recurrence rate. In pediatric patients with hemoptysis, comprehensive intraprocedural angiography is essential to identify all culprit feeding arteries. Furthermore, the NBCA mixture ratio must be precisely adjusted based on shunt volume and blood flow velocity to achieve complete occlusion of the shunt, thereby reducing the risk of postprocedural recurrence.

Abbreviations

SA-PA

systemic artery-to-pulmonary artery

NBCA

n-butyl-2-cyanoacrylate glue

PVA

polyvinyl alcohol

DSA

digital subtraction angiography

IQR

interquartile range

SIR

Society of Interventional Radiology

Authors’ contributions

Z.S. Conceptualization, Methodology, Investigation, Formal Analysis, Data Curation, Writing – Original Draft Preparation, Writing – Review & Editing, Visualization.X.L. Conceptualization, Methodology, Resources, Writing – Review & Editing, Supervision, Project Administration.All authors read and approved the final manuscript.

Funding

This study did not receive any specific funding.

Data availability

The datasets used and/or analyzed during the current study are not publicly available due to patient privacy and confidentiality regulations but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study strictly followed the ethical principles of the World Medical Association Declaration of Helsinki (2024 revision), and the protocol was approved by the Ethics Committee of Children’s Hospital of Soochow University (Approval No. 2026CS068). This was a retrospective de-identified study with no additional diagnostic and treatment risks, and written informed consent from the guardians of the children was waived after approval by the Ethics Committee.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.