Flexible bronchoscopy (FB) is a technique with diagnostic and therapeutic applications that may be useful in patients with congenital heart admitted to the paediatric intensive care. Our aim was to describe the characteristics of patients that underwent FB, the indications for the procedure, its findings and associated complications.

We conducted a retrospective and descriptive analysis of 21 FB procedures performed in 15 cardiac patients in the paediatric intensive care unit of a tertiary care hospital between 2013 and 2019. The proportion of patients with heart defects that underwent FB was 1.9% of the total that underwent surgical correction. We reviewed the health records of the patients and collected data on demographic characteristics, heart disease, indications for FB, respiratory support, FB findings and complications.

Two patients underwent 2 FB procedures and 1 patient 5, the rest underwent a single procedure. Four patients (26.7%) died of causes not associated directly or indirectly with the procedure, while the overall in-hospital mortality after cardiac surgery in our centre was 3.6% in the same period.

Patients underwent FB at a median age of 18 months (2–35 months) with a median weight of 5870 g (3800−6600 g). The median age of patients that underwent cardiovascular surgery in the same period was 15 months (2–52 months) and the median weight was 8800 g (3800−16 000 g).

Table 1 summarises the modalities of respiratory support used during FB, with mechanical ventilation being most frequent (50%). Three patients received support with extracorporeal membrane oxygenation. The table also presents the indications for FB and the underlying diseases.

The most frequent findings of FB (Table 1) were left bronchial stenosis and malacia. Pathogens were isolated in 6 of the 8 bronchoalveolar lavage samples: Stenotrophomonas maltophilia in 2, methicillin-resistant Staphylococcus aureus in 1, Pseudomonas aeruginosa in 1 and Klebsiella pneumoniae in 1, in addition to a positive galactomannan antigen test in another patient (indicative of fungal infection).

In 9 procedures, a pulmonary toilet was performed to remove mucus plugs, and in 3, adrenaline was delivered to treat mild bleeding. Steroid therapy was indicated in 1 patient and positive airway pressure in another 3, with no change to management in any other patients (Table 1).

Transient complications developed in 2 procedures: mild bleeding in one, and desaturation in another that required intermittent withdrawal of the bronchoscope.

Congenital heart and airway diseases are interdependent conditions given their common embryonic origin.1 More than 3% of patients with congenital heart diseases1 have associated airway malformations that may prolong disease, increase the risk of respiratory tract infections, be life-threatening and hinder recovery.

Magnetic resonance imaging2,3 is a non-invasive technique considered useful for diagnosis of airway malformations, but it carries risks for patients who are haemodynamically unstable, especially those who require invasive ventilation, as was the case in 50% of the sample under study. Thus, bedside FB could be a better option. In 11 of the 21 procedures (52%), the findings of FB led to changes in management in the form of prescription of treatments not previously used or the use of FB itself for treatment purposes. In 7 of the 10 procedures that did not lead to changes in management, a watchful waiting approach was chosen because the findings included areas of malacia that, given the age of the patients, were expected to resolve with adequate development of the airway through natural growth and maturation without invasive intervention, which has since been corroborated by none of them requiring surgical intervention in the airway at a later time.

Flexible bronchoscopy is a useful and safe technique2,4,5 for the dynamic assessment of the airway that also allows performance of pulmonary hygiene interventions, delivery of substances and obtention of bronchoalveolar lavage samples that may guide antibiotherapy during the procedure. Recurrent or persistent atelectasis is the most frequent indication for FB in these patients.

The spectrum of airway disorders in our sample was similar to those described in other studies2: extrinsic compression of the airway by cardiac or vascular structures are the most frequent cause of obstruction, in addition to obstruction due to mucous secretions.6 The left main bronchus is involved most frequently due to its position relative to the left atrium and the homolateral pulmonary artery. Between 13% and 26% of vascular malformations, depending on the series,2 cause compression of the bronchial tree. Another aspect to consider is the possible presence of subglottic and tracheal stenosis after surgery5 secondary to prolonged intubation and inflammation.

The detection of these abnormalities is useful for patient management, as surgical correction or positive airway pressure may be considered in the case of extrinsic compression. On the other hand, in the case of malacia, positive airway pressure is the temporary treatment of choice, as the problem resolves with growth in nearly all patients.

There may be complications of FB,4,7 in most cases associated with risk factors in the patient, such as bleeding, fever, desaturation, bronchospasm or pneumothorax. Most are mild and self-limited, and the incidence of complications in our case series was 10%, which was consistent with the previous literature.

On one hand, it is important to consider that patients with operated heart defects that undergo FB tend to weigh less and are usually younger than patients with operated heart defects that do not require this procedure. This, along with other comorbidities, could have played a role in the greater mortality observed in patients that underwent FB, and this procedure not have been the direct cause of the increase, as these patients were in poorer condition or had more severe disease at the time of surgical intervention.

The main limitations of the study, in addition to its retrospective design, are its small sample size, unavoidable given that it was conducted in a single centre, and few hospitals in Spain—to our knowledge—have published case series on the subject.

Thus, to conclude, we believe that FB may offer a high diagnostic and therapeutic yield in neonates and infants with surgically corrected congenital heart defects, and while considered an invasive technique and precisely on this account, its potential benefits and risk should be assessed on a case-by-case basis before its performance. Expert use of FB and its performance based on shared decision-making by intensive care physicians, cardiologists and bronchoscopy practitioners make this technique reasonably safe, but it is never without risk or a routine procedure. It is also important to remember potential complications, in which the underlying disease and condition of the patient at the time of the procedure play a significant role.