Results
A total of 42 infants were enrolled (Table 1). Thirty-two (76.2%) infants were transferred from other medical centers. The mean gestational age at birth was 27.1 weeks. At the time of FE, infants had median PMA of 43.2 weeks, median chronological age of 15.1 weeks, and mean (SD) body weight of 3.4 (0.6) kg. Of the 22/42 (52.4%) infants who required respiratory supports with invasive airway, 15/42 (35.7%) had ETs and 5/42 (11.9%) had tracheostomy tubes; PPV with nasal prongs was used in 16/42 (38.1%) infants and oxygen cannula was used in 4 (9.5%) infants. The leading indications for FE assessment were failure to wean PPV in 36 (85.7%), failed ET extubation in 27 (64.3%) and abnormal breathing sound in 15 (35.7%). Thirty-five (83.3%) infants had more than one indication.
Endoscopes used
Two thin flexible endoscopies without inner channel, Olympus LFP scope (out diameter 1.8 mm, working length 60 cm) or Olympus LFP scope (out diameter 2.6 mm, working length 30 cm), were used for FE-NIV procedures. Endoscopes of appropriate size were chosen to pass and assess through 2.5 mm to 4.0 mm inner diameters of invasive airway as well as from the nose to all approachable AE tract in infants without invasive airway.
FE revealed AE tracts pathologies in 38 (90.5%) infants where total of 129 lesions were identified (Table 2). There included 111 airway lesions in 35 (83.3%) infants and 18 esophageal problems in 15 (35.7%) infants. Among them, 28 (66.7%) infants had detected more than one lesion. Of the 111 airway lesions, most frequently identified were bronchial granulations (28, 25.2%), TMs (18, 16.2%), BMs (15, 13.5%) and subglottic stenoses (11, 9.9%). Bronchial granulations were right side dominant (right 21, left 10 and bilateral 7), and all at distal bifurcations of the primary bronchi, contributing to local malacia. The 18 esophageal lesions identified comprised of 12 (66.7%) loose of gastroesophageal junction, 3 (16.7%) esophageal inlet stenoses and 3 (16.7%) mid-esophageal stenoses.
The mean (SD) duration of FE-NIV procedure for infant with invasive airway was 102 (33) seconds, whereas for infants without airway tubing, assessing whole three routes of AE tract was 227 (55) seconds. (Table 3) Eight (19%) infants had developed transient desaturation (<85% or less than 10% of prior baseline level) or bradycardia (<100 beat/min) more than than 60 seconds. They all got recovery within 60 seconds after a brief NIV (NC-AC) maneuvers. Durations of desaturation or bradycardia were all less than 120 seconds. No pulmonary air-leak associated with the FE-NIV procedures was noted. All infants successfully withstood and completed the FE-NIV procedures.
FE findings led to resultant changes in clinical management in 38/42 (90.5%) infants (Table 4), where a total of 158 changes along with the respective rationales was tallied. There were 64 changes of respiratory care among 36/42 (85.7%) infants, which consisted of titrations of PPV (29/64, 45.3%), shallow suctioning (17/64, 26.6%), ET (or tracheostomy tube) insertion depth adjustments (10/64, 15.6%) and ET extubations (8/64, 12.5%). Twenty-one (21/42, 50.0%) infants had changed 50 medications which included addition of systemic steroids (15/50, 30.0%), addition of anti-reflux medicine (14/50, 28.0%), discontinuation of antibiotics (13/50, 26.0%), and addition of antibiotics (8/50, 16.0%). In 18 (42.9%) infants, 37 therapeutic interventions were performed with the aid of FE-NIV, which included balloon dilatations of narrow lumens (14/37, 37.8%), laser-plasty for stenotic lesions or granulation ablations (13/37, 35.1%) and stents implantation (10/37, 27.0%) for severe (collapse more than 90%) tracheobronchial malacia caused difficult weaning the PPV. In these 10 stents implantations, there were 4 TM, 4 CM (2 right and 2 left) and 2 left main BM. Seven (7/42, 16.7%) infants received surgical procedures due to failure of medical management which included 4 tracheostomies for severe subglottic stenosis and 3 fundoplications for severe GERD.