342 - Patterns of Cardiorespiratory Events (CREs) Post-Extubation in Extremely Preterm Infants

Publication Number: 2331.342

Lina M. Ikhlef, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada; Ana Gabriela Saavedra Ruiz, RI-MUHC, Laval, PQ, Canada; Marc Beltempo, McGill University, Montreal, PQ, Canada; Robert Kearney, McGill University Faculty of Medicine and Health Sciences, Montreal, PQ, Canada; Guilherme Santanna, McGill University Health Center, MONTREAL, PQ, Canada; Wissam Shalish, McGill University Health Center, Montreal, PQ, Canada

Background: Respiratory pauses, desaturations, and bradycardias frequently occur in extremely preterm infants post-extubation, leading to clinical instability and need for reintubation.

Objective: (1) Describe the interrelationships of respiratory pauses, bradycardias, and desaturations during the immediate post-extubation period; (2) Assess for differences in CRE patterns between infants with successful vs failed extubation (reintubation within 7 days).

Design/Methods: Secondary, retrospective analysis of a crossover trial at the Montreal Children's Hospital NICU (2016-2018), including infants ≤1250g. Measures included breathing patterns from respiratory inductive plethysmography, heart rate (HR) from electrocardiography, and oxygen saturation (SpO2) from pulse oximetry for 2h immediately post-extubation. Desaturations were defined as SpO2 < 90% >2s and bradycardias as HR < 100bpm >5s. Respiratory pauses > 3s were identified using an Automated Unsupervised Respiratory Event Analysis algorithm. Event frequency, duration, transition probability, and timing were summarized and compared between infants with successful and failed extubation using nonparametric statistical tests.

Results: 56.7h of monitoring (mean 2.1h/infant) were obtained from 27 infants (26 WGA [25.4, 26.4]; birth weight 770g [610, 915]), during which 1174 respiratory pauses, 834 desaturations, and 54 bradycardias were recorded (Fig 1). Median (IQR) durations were 4.9s (3.7, 8.2) for pauses, 59.7s (31.5, 94.5) for desaturations, and 6.6s (5.8, 9.9) for bradycardias. Most events occurred in isolation (respiratory pauses: 96%; desaturations: 82%; bradycardias: 89%). Complex event patterns included respiratory pauses transitioning into desaturations (3.4%), desaturations transitioning into respiratory pauses (16.7%), and bradycardias transitioning into respiratory pauses (7.4%). The likelihood of a desaturation or bradycardia following respiratory pause increased with pause duration (Fig 2). Reintubated infants had a greater number of respiratory pauses >10s (p=0.01). Other parameters did not differ significantly between groups (Table 1).

Conclusion(s): During the first 2h post-extubation, brief respiratory pauses, bradycardias, and desaturations were common but mostly isolated. Longer pauses increased the likelihood of desaturation or bradycardia, and reintubated infants had more pauses >10s than those successfully extubated. These findings suggest that respiratory pause dynamics may have predictive value for cardiorespiratory instability and reintubation risk, a relationship that warrants further investigation in a larger cohort of extremely preterm infants.

Relationship between respiratory pauses, bradycardias, and desaturations.
Diagram illustrating the number of events and their overlaps. The total counts of each event type are shown inside rounded boxes (respiratory pauses: 1174, desaturations: 834, bradycardias: 54). Arrows indicate the number [and percentage] of overlapping events between categories (e.g., 40 pauses followed by desaturations, 139 desaturations followed by pauses, etc.).

Cardiorespiratory Event Frequencies, Durations, Transition Probabilities and Time Differences.
table 1.pdfThis table summarizes cardiorespiratory event characteristics for the overall cohort and stratified by extubation outcome.
i. Cardiorespiratory Event Frequency: The median (IQR) number of respiratory pauses (>3 s and >10 s), desaturations, and bradycardias observed per patient.
ii. Cardiorespiratory Event Duration: The median (IQR) duration (in seconds) of each event type.
iii. Cardiorespiratory Event Transition Probability: The likelihood of one event type leading to another (e.g., respiratory pause → desaturation).
iv. Cardiorespiratory Event Time Differences: The median time (in seconds) between sequential events, measured from the start of the first event to the start of the second event.
P-values represent comparisons between successful and failed extubation groups. Statistical significance was defined as p < 0.05.

Probability of desaturation or bradycardia following a respiratory pause by pause duration.
The probability of a subsequent desaturation (green line) or bradycardia (blue line) is plotted as a function of respiratory pause duration (seconds). Shaded areas represent 95% confidence intervals (Wilson score method).