OBJECTIVE:
To identify esophageal sensitivity phenotypes relative to acid (S
Acid
), bolus (SBolus), acid and bolus (S
Acid+Bolus
), and none (S
None
) exposures in infants suspected with gastroesophageal reflux disease (GERD).
METHODS:
Symptomatic infants (N=279) were evaluated for GERD at 42(40–45) weeks postmenstrual age using 24-hour pH-impedance. Symptom associated probability (SAP) for acid and bolus components defined esophageal sensitivity: 1) S
Acid
as SAP≥95% for acid (pH<4), 2) S
Bolus
as SAP≥95% for bolus, 3) S
Acid+Bolus
as SAP≥95% for acid and bolus, or 4) S
None
as SAP<95% for acid and bolus.
RESULTS:
Esophageal sensitivity prevalence (S
Acid
, SBolus, SAcid+Bolus, S
None
) was 28(10%), 94(34%), 65(23%), and 92(33%) respectively.
Emesis
occured more in SBolus and S
Acid+Bolus
vs S
None
(p<0.05). Magnitude (#/day) of
cough
and
emesis
events increased with S
Bolus
and S
Acid+Bolu
s vs S
None
(p<0.05). S
Acid+Bolus
had increased acid exposure vs S
None
(p<0.05). Distributions of feeding and breathing methods were distinct in infants with S
Bolus
vs S
None
(both, p<0.05). Multivariate analysis revealed that
arching and irritability events/day
were lesser at higher PMAs (p<0.001), greater for infants on NCPAP (p<0.01), with S
Bolus
and S
Acid+Bolu
s (p<0.05).
Coughs/day
was greater at higher PMAs (p<0.001), for infants with gavage and transitional feeding methods (p<0.02), with S
Bolu
s and S
Acid+Bolus
(p<0.05) but lesser with Trach (p<0.001).
Number of emesis events/day
were greater with SBolus and SAcid+Bolus (p<0.001).
Sneezes/day
decreased for infants on Trach (p=0.02).
CONCLUSIONS:
Feeding and breathing methods can influence the frequency and type of aerodigestive symptoms. We differentiated esophageal sensitivity phenotypes in NICU infants referred for GERD symptoms using pH-Impedance. Acid sensitivity alone was rare, which may explain poor response to acid suppressives; aerodigestive symptoms were predominantly linked with bolus spread. Magnitude of esophageal acid exposure and esophageal sensitivity to bolus spread may explain the pathophysiological basis for symptoms.