4 | DISCUSSION
The use of CPAP to treat RDS in premature infants has been associated with a lower incidence of chronic lung disease than invasive mechanical ventilation [13]. Therefore, there has been a notable shift to non-invasive respiratory support for these infants, and the term “less may be more” has been fully embraced in neonatology. However, increased use of CPAP requires a closer examination of its potential side-effects. The presence of an oro- or nasogastric tube in many of these infants may allow the delivered and swallowed gas to move to the stomach and further down the GIT. Abdominal distention may restrict diaphragmatic movement and lung expansion, and serious complications such as intestinal perforation and necrotizing enterocolitis have been reported [14,15,16]. Even though we analyzed the incidence of NEC and intestinal perforation, along with BPD, and found no significant difference between MD-nCPAP and bCPAP groups, our study was not powered to detect these differences ( Table 2) .
Studies using scintigraphy to determine the rate of gastric emptying in healthy term infants have yielded inconsistent results. While some studies have reported faster gastric emptying in older infants, a large recent study found that infants less than 3 months of age have faster gastric emptying than older infants and children [17,18,19]. As expected, gastric emptying is twice as fast in infants fed breast milk compared to formula [20]. Conversely, delayed gastric emptying has been associated with gastro-esophageal reflux in older children [21]. Using scintigraphy in VLBW infants, Gounaris et al. reported faster gastric emptying in those treated with MD-nCPAP compared to healthy controls [5]. However, the utility of scintigraphy is limited in children and infants because of the need to limit radiation exposure. US provides a noninvasive method to estimate the gastric emptying rate and can be performed at the patient’s bedside [22,23].
We found that neither the ACSA nor spheroid method gastric emptying rates were statistically different between infants treated with MD-nCPAP and bCPAP, analyzed by multivariable analysis adjusting for emptying phase (early, late) and GA (25-28 weeks, 28-34 weeks). The caloric density of the feedings, type of milk (expressed human milk only, donor human milk only, formula only, or mixed) or volume of milk were not significantly associated with the ACSA or spheroid rates. Corrected gestational age (CGA) was significantly different between the MD-nCPAP and bCPAP infants, but regression analysis of ACSA or spheroid rates and CGA showed no correlation between the two variables. Our findings are consistent with a recent meta-analysis that demonstrated no correlation between postmenstrual age and gastric emptying time [24].
A significantly faster gastric emptying was noted in the early phase (1-2 hours from the initiation of a feed) compared to the late phase (2-3 hours after initiation of a feed) when using the spheroid method, but this was seen only in the 25+0 to 27+6 week GA group using the ACSA method. This is consistent with previous studies showing faster gastric half-emptying times in the immediate post-prandial period using different methods of measurement [5,8,19]. A meta-analysis of 66 studies in various age groups performed by Bonner et al. revealed a non-linear emptying rate with faster early phase, especially for liquid-fed subjects compared to those fed solid food [24]. This can be explained by a high intragastric volume in the early phase, which stimulates the mucosal stretch receptors [29,30,31,32].
We noted high gastric residual percentages using both methods – 41% (30.3-49.8) by ACSA and 24.5% (16.8-32.3) by spheroid. Higher gastric residual percent measured by the ACSA method compared to the spheroid method is likely due to the antrum being the last part of the stomach to empty. It appears that traditional measurements of pre-feeding residual volumes by gastric aspiration underestimate the proportion of retained feeding by 19-25%. This may be dependent on patient position, feeding tube size, position of the tip of the tube, milk viscosity and aspiration technique [25,26,28]. Our subjects at the time of the US were receiving only non-invasive respiratory support and tolerating full enteral feeding without clinical evidence of feeding intolerance. This finding brings into question the reliability and value of using aspirated gastric residuals to assess feeding intolerance or gastrointestinal pathology in neonates [27,28].