Feeding intolerance, a common problem for preterm infants, is characterised by signs of delayed gastric emptying such as large gastric aspirates, abdominal distension, and vomiting. These signs may delay the progression of enteral feeding, the effect on the provision of adequate nutrition, and the development of successful oral feeding. Feeding intolerance is associated with necrotising enterocolitis, a gastrointestinal disease with high morbidity and mortality rates (1). Present methods for evaluating gastric emptying in the neonatal setting are limited to the assessment of gastric residual volumes through aspiration of the intragastric tube and sequential girth measurements for the evaluation of abdominal distension. Alternative methods of assessing gastric emptying reported in the literature have not been adopted in the clinical setting for many reasons (2–5).
The use of marker dilution techniques and scintigraphic methods has been researched in preterm infants (2,3), and neither is suitable for routine clinical use. Marker dilution provides a gross measure of gastric emptying by adding a known quantity of a dye marker to a “standard feed,” then aspirating, measuring, and replacing the stomach contents at single or multiple time points. This repeated aspiration and replacement of stomach contents are invasive and affect normal physiological processes. (6) Scintigraphic methods such as the 13C-octanoic acid breath test require exposure to ionising radiation, and are ideally performed after administering a “standard meal” of a set volume and composition; this is not appropriate for preterm infants who are prescribed specific individual feed volumes of varying compositions (5). In addition, these techniques require complex analyses to estimate gastric emptying (2,7). Alternatively, ultrasound offers a more physiological and noninvasive technique and has been used successfully to investigate gastric emptying in preterm infants, with antral cross-sectional area (ACSA) shown to correlate with feed volumes delivered to preterm infants (8,9).
Published studies report estimated gastric half-emptying times, using the time taken for the ACSA to reduce to half of its maximum area to determine the rate of emptying (9,10). The ACSA technique is limited in that it provides a proxy measure of gastric volume, allowing only for indirect estimates of gastric emptying patterns over time with no indication of actual gastric volumes. The ability to determine gastric volume would enable more sensitive measures of gastric emptying patterns as well as evaluation of gastric residuals in the preterm infant.
Lambrecht et al (11) demonstrated direct measurement of gastric volume in young infants using ultrasound imaging, but did not validate the technique for the preterm infant population. Lambrecht method is based on the assumption that the stomach is spheroid in shape. Observations of the fetal stomach in utero (12) and of the stillborn preterm infant's stomach by the first author at perinatal autopsies indicate that the stomach shape of a preterm infant may be fusiform or an irregular curved spheroid shape. It may be possible to increase the accuracy of existing methods of calculating stomach volume by using combinations of simple geometric shapes to more closely reflect the observed shape of the preterm infant's stomach.
Ultrasound examination of the stomach provides the opportunity to observe the characteristics of the ingesta as well as its volume and shape. The bright echogenic matter associated with milk feeds in infants is thought to indicate curd or precipitate, and may differ in volume and density between infants and between feeds of differing compositions, although this has not been investigated for breast milk feeds in preterm infants (13).
As for most biological measurements, gastric emptying has an intrinsic variability that influences the reliability and validity of measures. Previous studies of both the ACSA and direct stomach measurement techniques have typically assessed the reliability of the measures used, but not validated these against known volumes such as the feed volume delivered (8,11).
This study aimed to determine whether sonography is a reliable and reproducible method for determining gastric volume in preterm infants. Specifically, we compared 4 methods of estimating stomach volume using of a set of calculations based on measurements of sonographic images of the gastric antrum (ACSA) or of the stomach (stomach as spheroid, hemisphere + cylinder, and hemisphere + cone). Calculations were validated against the actual feed volume delivered to determine which method was most accurate. A secondary aim was to describe the echogenic characteristics of breast milk in the preterm infant's stomach during the course of delivery of an intragastric tube feed.
Medically stable preterm infants born at 28 to 34 weeks’ gestation who were receiving full enteral feeds via intragastric tube were recruited from the Special Care Nurseries of King Edward Memorial Hospital, Western Australia. Infants with congenital abnormalities, gastrointestinal disease, or symptoms of feeding intolerance within the previous 24 hours were excluded from the study.
To account for individual variability, it was determined that a minimum of 20 infants should be recruited. Recruitment continued until usable ACSA, and stomach images were available for 20 infants.
The ethics committees of the Women and Newborn Health Service and the University of Western Australia gave ethical approval for the study. Parents were provided with verbal and written information about the study and provided signed informed consent for their infants’ participation in the study.
Each infant was scanned during 1 feed (duration 14–35 minutes), at a time negotiated with parents and staff so as not to disrupt plans for breast-feeding and parent–infant contact. Scanning was scheduled to coincide with the infant's usual feed time, and was performed at the bedside in the neonatal nurseries using a portable ultrasound machine. After nursing observations and infant cares were attended, the infant was placed in the supine position and the shirt and bedding were loosened to enable access to the chest and abdomen. Sterile gel was warmed for infants in incubators, and used at room temperature for infants in open cots. The intragastric tube was aspirated and tested for acidity as per standard practice to confirm correct tube placement. Feeds were commenced immediately following the prefeed scan, with infants remaining in the supine position for the duration of the feed. All of the infants maintained their temperature within normal limits and tolerated the procedure well.
Serial images of the antrum and stomach were recorded immediately before commencement of the feed (0%), and during interruptions to feed delivery when 50%, 75%, and 100% of the total volume of the feed had been delivered. Feeds were delivered by gravity, and scanning duration was <3 minutes at each time point.
All studies were performed using real-time 2-dimensional ultrasound using portable ultrasound (Titan hand-carried Sonosite, Belrose, Australia) with the C11 paediatric transducer. To image the ACSA, the transducer was positioned perpendicular to the abdominal wall, inferior to the cipher-sternum parallel to the abdominal aorta at the level of the superior mesenteric artery, providing a transverse view of the antrum (9). To acquire images for calculation of stomach volume, the transducer was positioned perpendicular to the left lateral chest wall and rotated so that the stomach was both at its maximum diameter and length (longitudinal plane), and a clear view of the spleen was achieved. A second image was acquired with the transducer perpendicular to the longitudinal plane providing a transverse view (Fig. 1). The intercostal approach was most commonly used when scanning the stomach of the preterm infants because of the high position of the stomach. The anterior approach was only successful when there was a large volume of milk in the stomach. The position of the transducer was altered as necessary to compensate for infant movement. Only light pressure was exerted to avoid disturbing the infant. Images were recorded on the ultrasound machine and transferred to a laptop computer for measurement.
Ultrasound Image Analysis
One expert rater and 1 novice rater conducted image analyses. Before analysis, the raters selected the images to be measured; 1 each of the antrum and the transverse and longitudinal planes of the stomach for each time point (12 images per infant).
Maximum length of the longitudinal stomach image was measured and transverse and anterior-posterior measurements were made of the transverse stomach image (Fig. 1). Transverse and anterior-posterior measurements were made of the antral image. Each rater made duplicate measurements without access to their previous measurements or those of the other rater. The images were not measured in sequential order. For each infant, all measurements were made in a single session by each rater.
Measurements were made using Screen Calipers, version 4.0 (Iconico, New York, NY) on the raters’ individual laptop computers. These provide several advantages over traditional on-screen callipers, including the ability to use several callipers simultaneously, float callipers on the screen, lock rotational angles, and stay visible when changing images. This method has previously been used to reliably measure ultrasound images of nipple diameter during breast-feeding (14).
Echogenicity of Stomach Contents
Echogenicity of the stomach was assessed by rating the brightness of the longitudinal view of the stomach relative to that of the spleen. Images were classified as having “higher” or “lower” echogenicity if this was consistent across the stomach. Two classifications of variable echogenicity were used. “Snowstorm” referred to images with a predominantly fluid-filled stomach (anechoic) with many small bright specks interspersed in the fluid. “Mixed” referred to stomach contents appearing as a combination of lower and higher echogenic areas, indicating a mixture of fluid and semisolid contents (Fig. 2). Air was not imaged within the infants’ stomachs.
Areas of brightness within the stomach associated with milk feeds were assumed to indicate curd or precipitate. These more echogenic areas were further classified with respect to both density and volume. Density was classified as “high” where the bright echogenic areas appeared bright white (greater than the echogenicity of the spleen), and “low” where they were uneven or grey (equal or less echogenic compared to the spleen). Volume of the bright echogenic area was classified as “high” where it occupied more than half of the stomach area, and low otherwise. Ratings were made independently by each rater, once only for each longitudinal image, when the first set of measurements were taken.
Measures of gastric volume, including the (proxy) measurement of ACSA, and 3 different direct stomach volume calculations were tested using the following calculations:
Stomach Volume Measurements
- Spheroid (11): longitudinal × antero-posterior × transverse × 0.52
- Hemisphere + cylinder: (4/3πradius3)/2 + (πradius2 hemisphere)
- Hemisphere + cone: π/12 × transverse2 (longitudinal + ½transverse)
The spheroid equation is a variation of the standard equation for volume of a spheroid, using diameters rather than radii. The other 2 equations are approximations to the observed shape of the preterm infant stomach, using a combination of regular solid-shape volumes. The equations have been simplified to use diameter measurements rather than radii.
All analyses were performed using R 2.9.0 for Mac OSX (15). Additional packages nlme (16), irr (17), and lattice (18) were used for linear mixed modelling, intraclass correlations (ICCs), and lattice plots, respectively.
Intra- and Interrater Reliability
Reliability of the continuous stomach measurement data was assessed. Comparison of rater responses was made graphically using the method described by Bland and Altman (19) and analytically using the ICC approach of McGraw and Wong (20). The Bland-Altman plots were examined for patterns of disagreement between measurements that related to the magnitude of the measurement.
Intrarater reliability was assessed using the duplicate measurements, and was calculated separately for each rater. Interrater reliability was calculated using the means of the duplicates for each rater. For stomach volumes, the mean of the previously calculated volumes was used. For each measure, measurement bias, ICC for agreement (2-way variance), and 95% confidence intervals (CI) were calculated. Reliability of the 2 raters was considered to differ if the 95% CIs did not overlap. An ICC >0.8 is considered acceptable for a research setting and >0.95 acceptable for use in the clinical setting. (21).
To avoid biasing the assessment of the reliability, images in which the stomach was completely contracted (stomach volume = 0 mL) were omitted from the validation of image measurements. Reliability of the categorical echogenicity and stomach contents measurements were assessed using the κ statistic to determine the level of nonchance agreement. For the echogenicity rating, which was the only variable with >2 classification levels, the level of exact agreement was assessed.
Patterns across the increments of the feed were assessed qualitatively for each of the 4 measures. Relations between the echogenicity of stomach contents and feed type were assessed for images taken immediately after delivery of the milk feed using the Pearson χ 2 proportion test.
Validation of Stomach Volume Techniques
Calculated direct and proxy measures of stomach volume were validated against delivered feed volumes at 50%, 75%, and 100% of feed by determining whether the discrepancies were significantly different from zero, and whether the feed proportion delivered influenced the error. Linear regression and linear mixed effects (with feed as the group and individual baseline errors) were considered as possible models. The linear mixed-effect model was only significant for the spheroid calculation, indicating discrepancies were not random.
The first measurements of the novice rater were used for the stomach volume calculations, allowing testing of worst case scenario (likely least accurate) measurements, and reflecting clinical practice where single measurements are most likely to be used. To determine whether measured prefeed residuals influenced the measured feed volumes at subsequent time points, linear mixed-effects model analysis was used.
Of the 25 infants recruited, there were 19 complete ACSA and 17 complete stomach datasets. Missing data consisted of 1 complete dataset lost because of computer failure; 5 antral images discarded because they were not taken in the correct plane; final ACSA image unusable for 1 infant; and stomach images at 75% and 100% of feed delivery discarded because of poor image quality for 1 infant. For 6 infants, the stomach was fully contracted, that is, no measurable volume immediately before feeding; therefore, the prefeed images were not included in calculation of prefeed residuals.
Characteristics of the infants (13 girls, 11 boys) included in the analysis were mean (range): birth gestation 31 weeks (28–34), birth weight 1620 g (1040–2340), and postnatal age 23 days (2–64). Eight (30%) were small for gestational age (birth weight <10th centile), with the remaining infants appropriately grown at birth.
All infants received breast milk feeds; 16 received their mother's own expressed breast milk and 8 received pasteurised donor human milk (PDHM). S-26 SMA human milk fortifier (Wyeth Nutrition, Baulkham Hills, NSW, Australia) was added to the feeds of 13 infants (7 mother's own breast milk, 6 PDHM) at the recommended dilution of 1 sachet to 50 mL of breast milk. The mean (range) study feed volume was 36 mL (12–55), and the mean prescribed daily feed volume was 157 mL · kg−1 · day−1 (96–184).
Intrarater Reliability of Stomach Volume Measurements
There was a high level of agreement between duplicate measurements performed by the 2 raters for both antero-posterior and transverse stomach measurements. Other than the antral transverse measurement with ICC >0.95, all measurements had ICCs >0.99 (95% CI) (Table 1). Means and standard deviations (SDs) of the differences between the duplicates performed by each rater are presented in Table 2. The average mean differences were extremely small—the most extreme was −0.18 mm for the transverse stomach measurement and −0.22 mm3 for one of the calculated volumes—with SDs higher for the expert rater. All intrarater ICC agreement values were within the recommended clinically acceptable threshold of >0.95 (22).
Interrater Reliability of Stomach Volume Measurements
Means and SDs of the differences between the measurements performed by each rater are presented in Table 3. Despite wide limits of agreement between raters’ measurements, there was no evidence of bias. The most extreme mean difference between final measurements was 0.89 mm (antrum antero-posterior, SD 2.49 mm).
No bias was seen in either the intra- or interrater comparisons. Thus, there was no tendency within raters for the first and second measurements to differ systematically, and no evidence of the raters measuring differently (Table 4).
Prefeed Stomach Residuals
Of the 24 infants with prefeed stomach images, 18 had measurable gastric residuals and 6 had completely contracted stomachs. Residual measurements varied between the 3 calculation methods used, with a median residual volume and range of 2.12 mL (0.56–9.27 mL) for spheroid, 2.55 mL (0.65–12.22 mL) for hemisphere + cylinder, and 1.36 mL (0.43–6.70 mL) for hemisphere + cone calculations. The P values for the effect of the calculated feed residual volume on the calculated feed volumes were >0.2 for each method, indicating that knowing and accounting for the stomach residual volumes <9 mL (using spheroid calculation) did not provide a more accurate assessment of the stomach volume. Based on these results, residual volumes were not accounted for in subsequent gastric volume measurements.
Relation Between Stomach Calculations and Delivered Milk Volume
Statistically significant relations were seen between the delivered feed volume and all stomach volume calculation methods, including the ACSA. Individual variations in measurements were observed over time for all techniques (Fig. 3), and the direct stomach volume calculation was sensitive to small changes in volume. Accounting for the proportion of the feed already taken and the prefeed residual, a significant relation was found between spheroid volume calculations and the known delivered volumes of the same feed (P < 0.001), with small discrepancies that were not influenced by feed duration or feed volume.
A significant relation was seen between hemisphere + cylinder volume calculations and the known delivered feed volumes (P = 0.001); however, this method had a tendency to underestimate feed volumes, with the discrepancies increasing as the proportion of feed delivered increased. No association was seen with the prefeed residual (P = 0.86).
A significant relation was found between the hemisphere + cone volume calculation and the delivered volume (P < 0.001); however, this method had a tendency to overestimate feed volumes, with the discrepancies increasing as the proportion of feed delivered increased.
There was a significant relation between the ACSA and the proportion of feed volume delivered (P < 0.001); however, much variation was observed between individual infants. Although 5 (20%) infants demonstrated the expected monotonic increase in ACSA over the duration of the feed, a further 5 (25%) demonstrated the largest ACSA immediately before feeding and 8 (40%) infants demonstrated the largest ACSA during feed delivery, with subsequent reduction as the delivered feed volume increased. Measures of reliability of the ACSA were not performed because for a proportion of individuals, the ACSA did not reflect the expected increase in stomach volume that would result from cumulative increases in feed volume during feed delivery (Fig. 3).
Differences in ratings of ultrasound image characteristics occurred for 30 of 192 (15.6%) images. Nonchance agreement (κ, 95% CI) was highest for echogenicity (0.920, 0.504–0.848; P < 0.001) and lowest for curd volume (0.755, 0.606–0.904; P < 0.001). Ratings for presence of curd and curd density both achieved a κ >0.8. These results suggest that interrater reliability was extremely high for ratings of echogenicity, and substantial for ratings of presence of curd, curd density, and curd volume.
Characterisation of Stomach Echogenicity
All 4 different classifications of echogenicity were observed in this study (Fig. 2), with lower echogenicity the least common finding (Table 5). Prefeed (residual) stomach images were frequently more echogenic than the spleen (13/18 = 72%) and image characteristics changed during the course of the feed (Table 5). At completion of the feed, higher echogenicity was seen most often with fortified milk (P = 0.006), that is, 7 of 12 (58%) fortified compared to 2 of 11 (18%) unfortified. Lower echogenicity was only associated with unfortified mother's own milk (P < 0.001), whereas snowstorm was more common in unfortified milk (80%, not significant) (Table 6).
This study found diagnostic ultrasound to be a reliable and accurate method of measuring gastric volume in the preterm infant. Although adequate intra- and interrater reliability was demonstrated for the ACSA and for the 3 direct stomach measurement methods, direct measurement of stomach volume as a spheroid was the most accurate approximation of stomach volume and also allowed for evaluation of stomach contents. This technique is therefore ideal for determining gastric emptying and curding patterns in response to different types of milk feeds as well as tracking gastric emptying in preterm infants experiencing difficulties in transitioning to full enteral feeds.
Ultrasound is a noninvasive, well-tolerated, reproducible, and reliable method of calculating stomach volume over time using the direct measurement of stomach volume (spheroid calculation). Presently, there are no published reports of testing for intra- and interrater reliability for the direct stomach volume calculation or ACSA method in preterm and term infant populations.
Lambrecht et al (11) performed a spheroid calculation of gastric volume with ultrasound, although they did not report validation of the measurement technique. We have confirmed that this measurement more consistently estimated stomach volume (Table 4, Fig. 3), suggesting that neither the cone + hemisphere nor hemisphere + cylinder model are appropriate models for preterm infants’ stomach shape (Table 3). The hemisphere + cone model overestimated stomach volumes, with the discrepancies increasing as the proportion of feed delivered increased, whereas the hemisphere + cylinder model consistently underestimated stomach volumes. Further analysis showed that the discrepancies for each of these calculations were not related to feed duration and therefore were not a consequence of gastric emptying during feed delivery.
Although we demonstrated a significant relation between the proportion of feed delivered and ACSA, we found that this was the most inconsistent method of tracking gastric emptying (Table 4, Fig. 3). Newell et al (9) previously reported validity of the ACSA method by calculating correlation coefficients for ACSA and volume delivered during the course of an intragastric tube feed (n = 8); as in our study, a strong linear relation was demonstrated and significant correlation found (r = 0.78, P < 0.001). Reproducibility of the ACSA method was determined by comparing a series of 10 antral measurements taken during 1 minute; the method demonstrated reproducibility in a single rater with a mean (standard error of the mean) coefficient variation of 7.7% (1.1). Small sample size may have prevented the detection of a sizeable proportion of anomalous patterns of change in ACSA during the duration of a feed, as we have demonstrated in our study. Furthermore, contraction of the antrum and passage of digesta through the antrum may not have been visualised and recorded during imaging periods of 1-minute duration. The ACSA is the most common method used in assessing gastric emptying in the preterm infant and it is technically simpler to perform than the direct stomach measurement. Both techniques are relatively quick with minimal disruption to the infant, although some have reported that imaging every 10 minutes may be considered excessive handling for the extremely preterm infant (8,9,23). Similar to Newell et al (9), we found that measurement of the ACSA is highly repeatable for a single rater, with ICC agreement values >0.95; however, the ACSA was highly variable and less consistent than the direct stomach measurement method. Rather than demonstrating the expected monotonic increase in ACSA during the duration of the feed, 13 of 24 (54%) infants demonstrated a decrease in ACSA at 1 point during the observation period (9,23). In addition, Newell et al (9) indicated anomalies in 3 of 8 (38%) infants; 1 infant had a reduction in antral area during feeding, whereas 2 demonstrated no change in antral area from prefeed to the time point when 25% of the feed volume had been delivered. These anomalous changes may occur because of wide fluctuations in ACSA when the antrum is full (24), a more rapid transit of feed through the stomach in individual infants, or it may be that the technique requires a plane that is not consistently reproducible or is subject to questionable technique. Differences in infant positioning during scanning, with Newell's subjects positioned in right lateral position and ours positioned supine, did not affect the detection of anomalous ACSA findings. Both studies have therefore consistently found individual atypical ACSA responses during feed delivery, suggesting ACSA is not a reliable proxy measure of gastric volume.
Infants routinely had an intragastric tube aspirate collected for testing of pH to confirm tube placement as per hospital policy. Although only a minimal aspirate is required for this test, in cases in which there is a volume of fluid in the stomach, the sample is easily aspirated and it is clear to the clinician that further fluid is available for aspiration. In our study, in every case, only a scant aspirate could be obtained (<0.2 mL), suggesting that the stomach was empty in all of the infants studied. Despite this, a measureable pre-feed residual was frequently detected by ultrasound (66%), with some infants having a considerable volume. For example, 1 infant had a calculated residual volume of 9 mL that represented 28% of the previous feed (Table 5). This is an important finding, particularly in the context of the management of feeding intolerance in the absence of reliable measures of gastric residuals. Given that 26% to 48% of neonatal gastric tubes are potentially incorrectly placed (25,26), an ultrasound scan of the stomach would quickly and reliably alert clinicians to the presence of significant gastric residual volume. This is analogous to the routine practice of using ultrasound to calculate urinary bladder residual volumes in patients following prostate or gynaecological surgery (27,28).
Assessment of the echogenicity of stomach contents showed variable appearances across the duration of intragastric feeds of breast milk. Interrater reliability for ratings of echogenic appearance, presence, and volume of curd were found to be adequate. On completion of feed delivery, there was a marked difference in the appearance of fortified (more echogenic, Fig. 2) and unfortified breast milk (snowstorm, Fig. 2). The type and composition of milk feeds have been shown to affect gastric emptying in preterm infants (29,30) and may therefore differ in echogenic appearance between different types of milk with regard to volume and density (13). Indeed, Khan (13) has shown that in term infants, echogenicity of postfeed stomach images is directly related to the casein concentration (P = 0.003) and whey:casein ratio (P = 0.01) of the milk feed, with the highest casein concentrations associated with mixed echogenicity images. Bovine-based human milk fortifier is of a large particle size that is not soluble and has a higher casein content than breast milk and therefore reflects ultrasound waves more than unfortified breast milk. Conversely, the echogenic appearance of unfortified milk indicates a higher aqueous component that may contribute to a more rapid rate of stomach emptying. There are conflicting reports of the effect of human milk fortifiers on gastric emptying rates, with some reporting no difference in emptying (23,31) and others reporting faster emptying rates for unfortified breast milk (8,32,33). These differences are likely because of the differing compositions of commercial human milk fortifiers. Classification of ultrasound images and their associations with milk type, in conjunction with analysis of gastric emptying patterns, may prove useful in understanding the mechanisms of gastric emptying patterns in preterm infants.
The direct stomach measurement method of calculating stomach volume offers an accurate, noninvasive technique for monitoring gastric emptying and for detecting and measuring gastric residuals in the preterm population. Use of this technique has the potential to detect and further our understanding of gastric emptying patterns and large stomach residual volumes in the preterm infant. Concomitantly, ultrasound assessment offers further advantages such as the ability to observe the echogenicity and density of stomach contents and to detect pathologies such as the presence of lactobezoars (34). Investigation of relations between gastric emptying and echogenicity of different types of milk feeds will assist in the evaluation and management of preterm enteral nutrition.
These study findings are limited to the preterm infants born between 28 and 34 weeks of gestation who are fed by intragastric tube. Expansion of this study to include serial monitoring of gastric emptying using the direct stomach volume calculation method and comparison of findings with the published literature would expand our knowledge of normal patterns of gastric curding and emptying in the preterm population over time. The strengths of this study include the number of time points during which gastric volume was measured, and the range of infant ages and feed volumes. The high levels of intra- and interrater measurement agreement indicate that the direct stomach volume calculation method has the potential to be adopted as a clinical diagnostic tool.
Direct ultrasound stomach measurement (spheroid) will provide a useful research tool in investigating gastric emptying in preterm infants. Identification of the presence and nature of residual volumes and intragastric curd may be useful in furthering our understanding of feeding intolerance and gastric emptying patterns in relation to different feed types.
A girl of 7 .... with increasingly severe pains, repeated loss of consciousness, and moreover with the vomiting of a fecal liquid died miserably in 2 days. On opening the body, the cecum intestinum (appendix) was narrowed and constricted ... the obstacle overflowing opened up itself an unusual route into the abdominal cavity, by a necrosis and perforation a little about the obstructed place ....
Jean Fernel (1497–1558) Universa Medicina, 1657
The consummate humanist in medicine, Jean Fernel (pictured) was an astronomer, mathematician, and physician who systematized medical thought in his Universa Medicina (1554), which was divided into 3 parts: physiology, pathology, and therapeutics. Fernel drew attention to many Galenic errors. He opposed severe bloodletting, and correlated the respiratory and gastrointestinal manifestations of the influenza as 1 disease. He described the postmortem cited above, but his description of iliac passion—“passion” as a term to express exquisite pain—is the earliest recorded diagnosis of premortem appendicitis. More than 300 years later, American Charles McBurney (1845–1913), chief surgeon at Roosevelt Hospital, in 1888 began novel studies in appendicitis. He described progression, presenting symptoms, and signs and defined the point of greatest tenderness—now called McBurney point.
—Contributed by Angel R. Colón, MD
1. Lin PW, Stoll BJ. Necrotising enterocolitis. Lancet
2. Bode S, Dreyer M, Greisen G. Gastric emptying
and small intestinal transit time in preterm
infants: a scintigraphic method. J Pediatr Gastroenterol Nutr
3. Mezzacappa D-S, Marques MA, Collares EF. Gastric emptying
in premature newborns with acute respiratory distress. J Pediatr Gastroenterol Nutr
4. Nour S, Mangnall YF, Dickson JA, et al. Applied potential tomography in the measurement of gastric emptying
in infants. J Pediatr Gastroenterol Nutr
5. van Wijk MP, Benninga MA, Dent J, et al. Effect of body position changes on postprandial gastroesophageal reflux and gastric emptying
in the healthy premature neonate. J Pediatr
2007; 151:585–590.90 e1-2.
6. Cavell B. Gastric emptying
infants. Acta Paediatr Scand
7. Pozler O, Neumann D, Vorisek V, et al. Development of gastric emptying
in premature infants. Use of the C-octanoic acid breath test. Nutrition
8. Ewer AK, Durbin GM, Morgan ME, et al. Gastric emptying
infants. Arch Dis Child Fetal Neonatal Ed
9. Newell SJ, Chapman S, Booth IW. Ultrasonic assessment of gastric emptying
in the preterm
infant. Arch Dis Child
1993; 69 (1 Spec No):32–36.
10. Riezzo G, Indrio F, Montagna O, et al. Gastric electrical activity and gastric emptying
newborns fed standard and hydrolysate formulas. J Pediatr Gastroenterol Nutr
11. Lambrecht L, Robberecht E, Deschynkel K, et al. Ultrasonic evaluation of gastric clearing in young infants. Pediatr Radiol
12. Nagata S, Koyanagi T, Fukushima S, et al. Change in the three-dimensional shape of the stomach in the developing human fetus. Early Hum Dev
13. Khan S. Short-Term Variations in Breastmilk Composition: Associations with Feeding Patterns and Gastric Emptying in Term Infants
. PhD thesis. Perth: University of Western Australia; 2012.
14. McClellan H, Sakalidis V, Hepworth A, et al. Validation of nipple diameter and tongue movement measurements with B-mode ultrasound
during breastfeeding. Ultrasound Med Biol
15. R Development Core Team. A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2009.
19. Bland J, Altman D. Statistics notes: validating scales and indexes. BMJ
20. McGraw K, Wong S. Forming inferences about some intraclass correlation coefficients. Psychol Methods
21. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics
22. Nunnally JC, Bernstein IH. Psychometric theory. New York:McGraw-Hill; 1994.
23. McClure RJ, Newell SJ. Effect of fortifying breast milk on gastric emptying
. Arch Dis Child Fetal Neonatal Ed
24. Pedersen JF. A modified sonographic technique for assessment of gastric emptying
of liquid. Acta Radiol
25. de Boer JC, Smit BJ, Mainous RO. Nasogastric tube position and intragastric air collection in a neonatal intensive care population. Adv Neonatal Care
26. Freeman D, Saxton V, Holberton J. A weight-based formula for the estimation of gastric tube insertion length in newborns. Adv Neonatal Care
27. Bodker B, Lose G. Postoperative urinary retention in gynecologic patients. Int Urogynecol J Pelvic Floor Dysfunct
28. Palese A, Buchini S, Deroma L, et al. The effectiveness of the ultrasound
bladder scanner in reducing urinary tract infections: a meta-analysis. J Clin Nurs
29. Indrio F, Riezzo G, Raimondi F, et al. Prebiotics improve gastric motility and gastric electrical activity in preterm
newborns. J Pediatr Gastroenterol Nutr
30. Miller MJ, Witherly SA, Clark DA. Casein: a milk protein with diverse biologic consequences. Proc Soc Exp Biol Med
31. Gathwala G, Shaw C, Shaw P, et al. Human milk fortification and gastric emptying
in the preterm
neonate. Int J Clin Pract
32. Ewer AK, Yu VY. Gastric emptying
in pre-term infants: the effect of breast milk fortifier. Acta Paediatr
33. Yigit S, Akgoz A, Memisoglu A, et al. Breast milk fortification: effect on gastric emptying
. J Matern Fetal Neonatal Med
34. Heinz-Erian P, Gassner I, Klein-Franke A, et al. Gastric lactobezoar—a rare disorder? Orphanet J Rare Dis
Keywords:© 2013 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,
gastric emptying; preterm; ultrasound; stomach volume