Strap To Reduce Apnea in Preterm Infants: A Randomized Crossover Trial

Premature infants commonly receive continuous positive airway pressure (CPAP) or non-invasive positive pressure ventilation (NIPPV) as respiratory support. A frequent and underappreciated problem during nasal CPAP/NIPPV is mouth opening, which disrupts the closed circuit needed for effective positive pressure transmission. This oral air leak reduces effective airway pressure, leads to loss of lung volume, increases work of breathing, and results in respiratory instability. Clinically, this manifests as intermittent hypoxemia (IH) - episodes of oxygen desaturation - and bradycardia.

IH is highly prevalent in preterm infants and carries significant clinical consequences. A secondary analysis of the Canadian Oxygen Trial found that IH episodes lasting more than one minute were independently associated with death or neurodevelopmental impairment at 18 months (RR 1.66; 95% CI 1.35-2.05). The Pre-Vent study (Ambalavanan et al., Am J Respir Crit Care Med. 2023;208:79-97) demonstrated that IH80 duration per event (IH80dpe) - the average duration of episodes with SpO₂ below 80% for at least 10 seconds - was the strongest single predictor of death or invasive mechanical ventilation at 40 weeks postmenstrual age (OR 1.86; 95% CI 1.27-2.71) and grade 3 BPD (OR 1.70; 95% CI 1.18-2.45), supporting its selection as the primary outcome of this trial.

A chin strap is a simple, commercially available soft textile device designed to maintain mouth closure during CPAP. Chin straps are used in many NICUs, including at UAB, as part of standard clinical practice. However, despite widespread empirical use, no randomized controlled trial has evaluated their physiologic impact in preterm infants. Available evidence is limited to small observational studies and adult data from obstructive sleep apnea populations.

Study Design

This is a single-center pilot randomized AB/BA crossover trial. Each enrolled infant serves as their own control, receiving both conditions - a chin strap period and a no-strap period - each lasting 24 hours, separated by a 1-hour washout. Infants are randomized 1:1 to either the AB sequence (chin strap first, then no strap) or the BA sequence (no strap first, then chin strap). Randomization uses sequentially numbered sealed opaque envelopes with permuted block sizes.

Because the effect of the chin strap on IH is expected to be immediate and reversible upon removal, the crossover design is biologically appropriate and reduces the sample size needed compared to a parallel design. The 1-hour washout allows stabilization between periods.

Interventions

During the chin strap period, infants wear the Neotech ChinStrap (Neotech Products LLC), a soft, NeoFoam-based device sized to head circumference, applied to maintain gentle mouth closure. Mouth closure is confirmed by nursing staff with no visible gap between the lips, documented at application and each hourly assessment. During the no-strap period, infants receive standard CPAP/NIPPV without a chin strap, identical to routine clinical care.

CPAP/NIPPV settings are maintained at pre-enrollment levels throughout both periods. A pressure change of ±1 cmH₂O is permitted if clinically needed and must be documented. A change of ≥2 cmH₂O constitutes a protocol deviation. The escalation criterion is FiO₂ requirement greater than 0.70 to maintain SpO₂ within target range.

Monitoring

Continuous physiologic monitoring is performed throughout the 48-hour study period and includes pulse oximetry, transcutaneous CO₂, heart rate, respiratory rate, and nasopharyngeal (NP) pressure. NP pressure is measured using an FDA-cleared nasopharyngeal catheter connected to the NAPA LP-15 Airway Pressure Monitor (DRW Medical LLC). NP pressure reflects effective CPAP pressure delivered to the airway; oral air leak reduces NP pressure, making it a sensitive indicator of CPAP delivery efficiency.

Nursing staff perform hourly assessments including: (a) Mouth Closure Index (MCI) using a 0-3 scale (0 = fully closed, no leak; 1 = small leak; 2 = large leak; 3 = wide open); (b) infant posture (prone vs. supine); and (c) chin strap position and skin integrity. MCI inter-rater reliability will be established prior to study initiation with a target Cohen's kappa ≥0.80.

Outcomes

The primary outcome is IH80dpe (seconds/event) per 24-hour study period. Secondary outcomes include IH80 frequency, IH85 and IH90 frequency and duration per event, time in SpO₂ target range (91-95%), bradycardia frequency, FiO₂, TcCO₂, NP pressure, MCI score, infant posture, respiratory rate, abdominal girth, and feeding intolerance.

Sample Size

Based on Pre-Vent data (IH80dpe 25 ± 6 seconds), a clinically meaningful reduction of 6 seconds (25 to 19 seconds, 24%), 90% power, and two-tailed α = 0.05, a minimum of 11 completers is required by paired t-test (n = (1.96 + 1.282)² × 6²/6² = 10.51). A target of 17 completers was selected for additional robustness. Accounting for 15% attrition, 20 infants will be enrolled.

Statistical Analysis

The primary analysis uses a paired t-test or Wilcoxon signed-rank test comparing IH80dpe between the chin strap and no-strap periods. A linear mixed-effects model with treatment, period, sequence, and carryover as fixed effects serves as a sensitivity analysis. Carryover will be formally tested; if significant (p < 0.10), only first-period data will be used. Secondary outcomes are analyzed using paired t-tests, Wilcoxon signed-rank tests, or McNemar's test as appropriate. MCI score and posture will be included as covariates in the multivariable model.