373 - Association between Gastroesophageal Reflux and Hypoxemia in Premature Infants with Bronchopulmonary Dysplasia

Publication Number: 1357.373

Desiree Gutmann, Stanford University School of Medicine, Palo Alto, CA, United States; Kathleen Gibbs, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Matthew J. Ryan, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States; Erik Jensen, Stanford University School of Medicine, Palo Alto, CA, United States

Background: Whether gastroesophageal reflux (GER) contributes to cardiorespiratory instability in preterm infants with bronchopulmonary dysplasia (BPD) is uncertain.

Objective: Measure the temporal association between objectively recorded GER and hypoxemic episodes in very preterm infants with BPD.

Design/Methods: Single-center prospective observational study of simultaneous 24-h esophageal pH multichannel intraluminal impedance (pH-MII) and oxygen saturation (SpO2) monitoring in 33 very preterm infants with grade 2-3 BPD receiving full tube feedings. Before testing, treating clinicians (attending MD, bedside RN) were asked (yes/no) whether they believed GER actively contributed to the participant's ongoing lung disease. The strength of association between GER and hypoxemic (SpO2 < 80%) episodes was quantified by symptom sensitivity index (SSI, proportion of GER events associated with hypoxemia; positive if >10%), symptom index (SI, proportion of hypoxemic events associated with GER; positive if >50%), and symptom association probability (SAP, probability temporal relationship between GER and hypoxemia not due to chance; positive if >95%) using simultaneous monitor outputs segmented into 2min intervals. Characteristics of infants with vs without positive results for all 3 measures were compared.

Results: Study infants (Table 1) experienced a median of 32 (range 1-143) GER events and 5.6% (range 0.8-29.7) of time in hypoxemia (Spearman's rho: -0.18, p=0.32). SSI was positive in 85% (28/33), SI in 21% (7/33), SAP in 39% (13/33); all 3 indices were positive in 18% (6/33) (Figure). Infants with vs without positive results for all 3 measures experienced more GER events by impedance (median, IQR: 104, 65-118 vs 27, 7-53; p=0.002) and greater exposure of the proximal esophagus to GER (median, IQR: 2.7min, 1.1-4.2 vs 0.2min, 0.03-1.1; p=0.006), but similar proportions of time in hypoxemia (Table 2). A modest but non-significantly higher proportion of clinicians deemed that GER actively contributed to lung disease in infants positive for all 3 symptom association measures (Table 2).

Conclusion(s): A minority (18%) of infants demonstrated a strong, consistent association between episodic GER and hypoxemia. This subset experienced a significantly greater burden of GER than other study participants, but a similar proportion of time in hypoxemia, suggesting that infants with BPD who demonstrate a strong correlation between GER and hypoxemia may not be at increased risk of episodic cardiorespiratory instability, but successful treatment of GER might decrease related hypoxemia. Objective testing is needed to identify such candidates.

Characteristics of Study Cohort
Data displayed as count data with percentage or median, and range. PMA = post menstrual age; IMV = invasive mechanical ventilation; CPAP = continuous positive airway pressure; HFNC = high flow nasal cannula; NC = nasal cannula; RA = room air; MBM = maternal breast milk.

Comparison of Infants With and Without Reflux/Hypoxemia Association
Infants with and without reflux/hypoxemia association displayed as count data with percentage or median and IQR. IMV = invasive mechanical ventilation; MBM = maternal breast milk.

Symptom Association Analysis
Displayed are the symptom sensitivity index (SSI), symptom index (SI), and symptom association probability (SAP) for each study participant. Individual indices considered positive (indicated by shaded boxes) when the SSI >10%, SI >50%, and SAP >95%.