Esophageal body and lower esophageal sphincter function in healthy premature infants

GASTROENTEROLOGY1995;109:1757-1764

Esophageal Body and Lower Esophageal Sphincter Function in Healthy Premature Infants TAHER I. OMARI,* KAZUNORI MIKI,* ROBERT FRASER,* GEOFF DAVIDSON,* ROSS HASLAM, g WENDY GOLDSWORTHY, ~ MALCOLM BAKEWELL, LI HISAYOSHI KAWAHARA, ~ and JOHN DENT* Departments of *Gastroenterology,§Neonatology,and iIBiomedicalEngineering,Women's and Children's Hospital, North Adelaide,Australia; *Department of GastrointestinalMedicine, RoyalAdelaide Hospital, Adelaide, Australia; and ~Departmentof Pediatric Surgery, Kure National Hospital, Hiroshima,Japan

Background & Aims: Gastroesophageal reflux is a common problem in premature infants. The aim of this study was to use a novel manometric technique to measure esophageal body and lower esophageal sphincter pressures in premature infants. Methods: Micromanometric feeding assemblies (OD, -<2 mm) incorporating 4 - 9 manometric channels were used in 49 studies of 27 premature neonates. Esophageal body motility was recorded at three sites for 20 minutes after feeding. Twenty attempts (one per minute) were made to stimulate swallowing via facial stimulation (Santmyer reflex). In 32 studies, lower esophageal sphincter pressures were recorded (sleeve) for 15 minutes before and after feeding. Results: Peristaltic motor patterns were less common than nonperistaltic motor patterns (26.6% vs. 73.4%; P < 0.0001) that comprised 31.1% synchronous, 34.6% incomplete, and 6.3% retrograde pressure waves. Reflex swallowing was elicited more frequently in neonates older than 34 weeks postconceptional age than in younger infants (33.4% vs. 20.4%; P < 0.05). Mean lower esophageal sphincter pressure was 20.5 + 1.7 mm Hg before and 13.7 +_ 1.3 mm Hg after feeding (P < 0.0005). Conclusions: Premature infants show nonperistaltic esophageal motility that may contribute to poor clearance of refluxed material. In contrast, the lower esophageal sphincter mechanisms seem well developed.

astroesophageal reflux (GER) disease is a common problem in infants that has been associated with significant morbidity. 1 Reflux episodes may trigger acute respiratory problems such as apnea by reflex airway obstruction and/or laryngospasm. 2'3 Chronic lung disease can also result from pulmonary aspiration. 4 GER has also been postulated as a possible cause of sudden infant death syndrome) In adults and older children, the lower esophageal sphincter (LES) is known to be the major barrier to reflux of gastric contents into the esophagus. Most episodes of GER result from transient relaxations of the LES, 6 but esophageal and extraesophageal complications of GER may also be exacerbated by reduced clearance of refluxate

G

from the esophagus because of esophageal body peristaltic dysfunction] In contrast to older children and adults, in premature neonates there is little information about the esophageal motor mechanisms responsible for competence at the gastroesophageal junction or subsequent clearance of refluxate. Limited manometric evaluations of the LES of premature infants by manometric assembly pull-through have found that basal LES pressure is < 6 m m Hg 8'9 and increases with maturity. *° The side hole technique used in these studies provides only a brief record of the highly variable LES pressure profile 6 and does not enable reliable evaluation of swallow-related LES relaxation due to movement of the sphincter relative to the side hole. *1a2 Reports about esophageal body peristalsis in premature infants are conflicting; one study has shown poorly coordinated 8 and another study predominantly coordinated esophageal contractions. .3 The relative paucity of data in premature babies is largely due to the lack of suitable measurement techniques. Standard manometric catheters have a diameter of 4 . 5 - 5 . 0 m m and require perfusion rates of 0,5 mLl rain per channel to obtain satisfactory pressure increase rates. Both the catheter size and volume of perfusate used in conventional manometry are clearly inappropriate for neonatal studies. A recent study from our laboratory has shown that manometric channels with diameters of 0.35 m m have acceptable fidelity and pressure increase rates with perfusion rates as low as 0.05 mL/min, 14 W e have now developed multilumen manometric extrusions ( 4 9 manometric channels + feeding channel) that have an external diameter of 2 m m or less. Further development of this technique has enabled us to incorporate a miniature sleeve sensor for monitoring LES motility. The aim of the current study was to use this new manometric Abbreviations used in this paper: GER, gastroesophagealreflux; PCA, postconceptualage; PNA, postnatal age. © 1995 by the AmericanGastroenterological Association 0016-5085/95/$3.00

1758

OMARI ET AL.

approach to examine LES and esophageal body function in normal premature infants before and after feeding and during swallowing stimulated by a p u f f of air to the face (Santmyer reflex). 15'16

Materials and Methods Subjects Forty-nine studies were performed in 27 preterm infants (14 boys and 13 girls) whose postconceptional age (PCA) ranged from 33 to 38 weeks. PCA was calculated by adding postnatal age (PNA) ( 1 - 1 2 weeks) to gestational age (25-35 weeks). Gestational age was determined at birth from both maternal history of due dates and ultrasonic intrauterine morphometric assessment. When there was a discrepancy between maternal history and ultrasound assessment, gestational age was determined by examining the anterior vascular capsule of the eye lens. 17 In addition to the PCA, PNA, and gestational age, the body weight, previous feeding history, and arousal state at the time of each study were also recorded. Infants were screened for suitability before the study by one of the investigators (R.H.). All infants were well at the time of the study, had no evidence of neurological dysfunction, and were not receiving prokinetic medication. Due to the frequent use of theophylline for the treatment of apnea and bradycardia, infants receiving theophylline (n = 11) were not excluded. All infants were undergoing routine gavage feeding with nonfortifled expressed breast milk or infant formula (Enfalac 20/24 calorie; Mead Johnson). The study protocol was approved by the ethics research committee of Adelaide Women's and Children's Hospital, and informed written parental consent was obtained before each study.

Recording Technique The manometric approach was devised to permit monitoring of pressures with minimal infusion and with an assembly that could also be used to feed the infants. In addition, the liquid pathways of the manometric system were designed so they could be sterilized to remove any possible infection hazards to the infants.

Perfusion System A modified pneumohydraulic perfusion technique capable of delivering very low flow rates was used in all studies. To avoid the dampening effects of bubble entrapment on the pressure increase rate, the perfusion system, including the water reservoir, was preflushed with CO2 before filling with sterile degassed distilled water. Any CO2 bubbles still retained in the system after filling with water dissolved into solution within 10 minutes. Perfusion rates were accurately regulated with purpose-built stainless-steel capillary restrictors (1D, 0.1 mm; OD, 1.5 mm) that were calibrated to deliver a flow rate of 0.05 mL/min at a fixed driving pressure of 400 mm Hg. At 0.05 mL/min, the manometric channels had an offset of 1 0 15 m m Hg due to internal resistance. The pressure increase

GASTROENTEROLOGY Vol. 109, No. 6

rates within manometric lumina were determined by cannulation of the side holes with a blunted needle and occlusion with a noncompliant tap to prevent shock waves. All manometric lumina showed pressure increase rates > 1 0 0 mm Hg/s (179.8 -+ 12.5 mm Hg/s). The performance of the miniature sleeve sensor was assessed using a miniaturized version of a sphincter model. I8 Previously described methods of testing in the model 18 showed uniformity of sleeve sensitivity along its length on pull-through. The pressure increase rate at a perfusion rate of 0.05 mL/min ranged along its length from 11.0 _+ 0.1 mm Hg/s (distal) to 55.7 + 0.8 mm Hg/s (proximal). Between each study, the catheter, manifold, and restrictors were sterilized either by autoclave or with ethylene oxide.

Manometric Catheters Two purpose-built silicone rubber manometric microassemblies were used in the studies. These were constructed from purpose-designed multiple lumen microextrusions (Figure 1A). The first 17 studies evaluated only esophageal body motility with a 1.7-mm diameter, 5-1umen assembly. This had three side holes, 1.5 cm apart, that were positioned in the esophageal body. The second assembly (OD, 2 mm), which was used in the subsequent 32 studies, contained nine recording lumina (ID, 0.35 mm) and incorporated a 2.5-cmlong sleeve sensor (Figure 1). Side holes positioned along the extrusion recorded pressures from the pharynx (two), esophageal body (three), LES (sleeve), and stomach (one) (Figure 1B). Pharyngeal side holes were water-filled but nonperfused, whereas esophageal, sleeve, and gastric side holes were perfused with sterile degassed H20 at 0.05 mL/min. In both assemblies, lumina were arranged around a larger feeding channel (Figure 1A). Swallowing was recorded with a water-filled pharyngeal side hole that was not perfused. Both manometric assemblies were positioned with the most distal esophageal body side hole approximately 1,5 cm above the LES high-pressure zone (Figure 1B). The nares-sphincter distance of the infants studied ranged from 14 to 17.5 cm. Analogue pressure transducer signals were amplified and filtered with a Synectics polygraph (Synectics, Stockholm, Sweden) and digitized at 10 Hz using an A-D card (National Instruments, Austin, TX). Data acquisition and analysis was performed on a Macintosh Quadra 700 with software based on National Instruments' Labview (M.A.D. software, C. Malbert, Royal Adelaide Hospital).

Protocol Two protocols were used in the study. In both, the assemblies were passed transnasally 30 minutes before feeds were due. In the first 17 studies, the 5-lumen assembly was introduced and positioned with the feeding lumen in the stomach. Following the feed, the catheter was withdrawn, and spontaneous patterns of esophageal motility were recorded for 20 minutes, during which an attempt was made each minute to stimulate Santmyer swallowing by administering standardized puffs of air to the face by using a neonatal compression bulb ventilator (Ambu-Bag).

December 1995

ESOPHAGEAL MOTILITY IN PRETERM NEONATES 1759

"

1.1

c

1mm

esophagua2 8o]

e'°pha~u'sem i T, ,°, ,\

o mm: g ~"

.*-4-l.[-I\\ e$ophagl.~$

"

-- --7~-

lmm

-

pe#used

~

]

__A___

LES O]

i lOseconds= infusion

Figure

1. (A) Cross sections of the 5-lumen (4 x 0.35 mm ID + 1 × 0.60 mm ID) and lO-lumen (9 x 0.35 mm ID + 1 x 0,75 mm ID) microextrusions used in the construction of the manometric assemblies. (B) Schematic diagram of the arrangement of side hole sensors for the sleeve assembly (lO-lumen extrusion) and its positioning within the esophagus of a premature infant. The alternative assembly (5-lumen extrusion) consisted of 3 side hole sensors only at locations corresponding to esophageal sensors 1-3. (C) An example tracing of pharyngeal, esophageal body, and LES motor patterns associated with swallowing in a 34-week (PCA) premature neonate.

Subsequent studies (n = 32) used the 9-lumen sleeve assembly. After a 15-minute accommodation period, LES pressure was recorded for 15 minutes immediately before and a f t e r feeding, and then spontaneous motility and Santmyer swallowing were assessed as previously described. To allow for the volume of distilled water infused for manometry during the study, the volume of the feed w a s reduced to 73% of normal. The feed volume ranged from 20 to 60 mL and was administered manually with a syringe during a 1 5 35-minute period.

Analysis of Manometric Tracings LES pressure was defined as the difference between end-expiratory LES pressure and gastric pressure. Mean endexpiratory LES pressure was determined by the computerized analysis program for each minute of the study and used to calculate the overall mean LES pressure for each recording period. Both primary (swallow related) and secondary (nonswallow related) esophageal pressure waves were classified together for the purpose of this analysis because there was no pharyngeal signal in the first 17 studies; in the remainder, the waterfilled pharyngeal side holes were not consistently effective in recording pharyngeal contractions. Esophageal body pressure waves were scored if their amplitudes were >--10 m m Hg. The patterns of esophageal body pressure waves were classified into four patterns of spontaneous esophageal body contraction: peristaltic, synchronous, incomplete, and retrograde. Pressure waves were only classified as peristaltic if they were observed in three esophageal body recording sites and maximally propagated in an aboral direction at <- 3 cm/s. Synchronous pressure

were scored when pressure waves occurred at the three esophageal sites and the maximum propagation velocity among channels was > 3 cm/s (in practice, synchronous pressure waves occurred simultaneously in multiple channels). Incomplete pressure wave sequences were scored if a wave was recorded at only a single side hole or waves were peristaltic or synchronous over only two recording sites. Retrograde pressure waves were defined as those propagated (<--3 cm/s) in an orad direction over either two or three recording sites. Successful stimulation of the Santmyer swallow reflex w a s taken as a peristaltic esophageal contraction that occurred between 2 and 4 seconds after administration of the puff of air. Air puffs that coincided with straining or occurred < 3 seconds a f t e r a spontaneous swallow were excluded from the analysis. The percentage success in eliciting a peristaltic response w a s calculated by division of the number of successful r e s p o n s e s by the number of attempts multiplied by 100. waves

Statistical Analysis Data are means (+SEM) unless otherwise stated. Paired data were compared using a Student's paired t test. Grouped d a t a of different sample sizes were compared using an F test and Scheffe's test. Correlation was determined using simple regression analysis. A P value of <0.05 was considered significant in all analyses.

Results T h e procedure was well tolerated b y all infants, and no adverse effects were noted related to the procedure. T h e infants' w e i g h t s ranged from 1390 to 3050 g (mean, 2118 g).

1760 OMARI ET AL.

A

Peristaltic

esophagus 2

Synchronous

Incomplete

T

eo1

'+'°io a ,+--T _!1 \ esophagus3

GASTROENTEROLOGYVol. 109, No, 6

so-[

I

/~

I

~

Retrograde

J ~T

B [ ] Peristaltic • Synchonous [] Incomplete • Retrograde [ ] Other

Figure 2. (A) Example tracings of the predominant esophageal motor patterns observed in the esophageal body of healthy premature infants. (B) The expression of the different motor patterns relative to the total number of pressure wave sequences recorded.

Esophageal Body Function

A total of 1819 pressure wave sequences were recorded. Four major patterns of pressure wave sequence were recorded from the esophageal body: peristaltic, synchronous, incomplete, and retrograde. Figure 2 shows these motor patterns and the proportion of pressure wave sequences that were grouped into each of the four categories. Nonperistaltic esophageal pressure waves (synchronous, incomplete, and retrograde) were three times more common than peristaltic esophageal pressure waves (73.4% vs. 26.6%; P < 0.0001). Only 25 pressure waves (1.4%) could not be adequately classified. The proportions of the different esophageal motor patterns according to PCA are shown in Figure 3. Age showed no major relationship to esophageal function because neither PCA nor PNA significantly influenced the proportion of peristaltic, synchronous, incomplete, or retrograde pressure waves (Figure 3A-D). Table 1 compares the average number of pressure wave types in each age group. Of these, the only change of statistical significance was a decrease in the average number of retrograde pressure waves in older infants (P < 0.05). Motility Associated With the Santmyer Reflex

Administration of a puff of air to the face induced swallowing in 26 of 27 infants, a response known as the Santmyer reflex. In the studies in which pharyngeal pressure waves generated by swallowing were recorded successfully, Santmyer swallows were manometrically

identical to primary swallows; i.e., a pharyngeal contraction followed 2 - 4 seconds later by an esophageal body pressure wave sequence (Figure 4A and B). When the puff of air failed to evoke an esophageal body response, this was either due to failure to propagate a swallow (Figure 4C) or failure to swallow (Figure 4D). Because we were not able to monitor swallowing in all babies, the expression of Santmyer swallowing was assessed solely on whether the stimulus evoked a peristaltic esophageal body response. In addition to swallowing, the puffs of air also elicited a range of responses including blinking, opening the eyes, and squirming. For each study, the percentage of swallows in response to the air puff varied between 0% and 73% with a mean of 28.9% - 2.9%. Figure 5 shows the effect of PCA on the proportion of successful Santmyer swallows. An increase in the percentage of successful Santmyer swallows was observed between 33 and 35 weeks PCA (20.4 ___ 3.5% for infants 34 weeks and younger vs. 33.4% + 3.8% for infants 35 weeks and older; P < 0.05). A weak but significant positive correlation (r = 0.29; P < 0.05) was found between increased expression of the reflex and PNA. Expression of Santmyer swallowing did not correlate with gestational age (r = 0.21) or body weight (r = 0.15). There was no significant difference in the expression of the reflex between infants with their eyes open (predominantly awake, 26.4% + 3.9%; n = 28) and infants with their eyes closed (probably sleeping, 31.5% -+ 4.3%; n = 17). LES Pressure

A high-pressure zone was present at the esophagogastric junction in all infants (Figure 1). In studies in which swallow-induced pharyngeal pressures were recorded successfully, the onset of LES relaxation occurred within +3 seconds of theupstroke of hypopharyngeal pressure. Preprandial LES pressure was higher than postprandial LES pressure (20.5 + 1.7 mm Hg vs. 13.7 --+ 1.3 mm Hg, respectively [P < 0.0005]). There was a trend for an increase in preprandial LES pressure with increasing PCA (Table 2), which did not achieve statistical significance (P = 0.12). Postprandial LES pressure did not change significantly with increasing PCA (Table 2). Discussion

In this study, we evaluated LES and esophageal body function in premature infants with a newly developed miniaturized manometric feeding assembly. Our recordings indicate that in healthy premature babies, more than 70% of esophageal body contractile sequences

December 1 9 9 5

ESOPHAGEAL MOTILITY IN PRETERM NEONATES

40

A

40

//

30 % OF TOTAL 10 0

33-34

35-36

30 % OF TOTAL

20 10 0

37-38

40

40

35-36

37-38

D

30

30 % OF TOTAL

20

20 10

10 0

33-34

POST-CONCEPTIONAL AGE (WEEKS)

POST-CONCEPTIONAL AGE (WEEKS)

% OF TOTAL

1761

33-34

35-36

37-38

POST-CONCEPTIONAL AGE (WEEKS)

0

33-34

35-36

37-38

POST-CONCEPTIONAL AGE (WEEKS)

Figme 3. The effect of PCA on the relative expression of different esophageal body motor patterns. Data expressed at mean _+ SEM ( 3 3 - 3 4 weeks, n = 17; 3 5 - 3 6 weeks, n = 19; 3 7 - 3 8 weeks, n = 13). (A) Peristaltic, (B) synchronous, (C) incomplete, and (D) retrograde.

are nonperistaltic. In addition, the motor mechanisms responsible for tonic contraction of the LES are well developed, and LES pressure seems to be sufficient to maintain esophagogastric competence. Few studies have examined LES and esophageal body motor patterns in premature infants because of the lack of a suitably sized assembly and concerns about the volume of perfusate that can be safely infused during manometry: These technical barriers have been overcome by the development of purpose-designed extrusions with 4 9 lumina that have the same external diameter as a stan-

Table 1. Type of Pressure Wave Sequence Age group 33-34 (n = 35-36 (n = 37-38 (n =

weeks 17) weeks 19) weeks 13)

Peristaltic

Synchronous

Incomplete

Retrograde

9.2 ± 1.0

8.6 ± 1.4

13.0 ± 1.9

3.5 ± 0.8

8.8 ± 1.0

11.9 _+ 1.6

11,2 _+ 1.4

2.0 ± 0.5

6,7 +_ 1.5

7.9 ± 1.4

9.2 +_ 1.3

1.1 ± 0.3 ~

NOTE. The average number of pressure wave sequences observed during a 20-minute postprandial period determined for different age groups of healthy premature infants. Data given as mean _+ SEM on n studies. aSignificantly different from the 33-34-week age group (P < 0.05).

dard neonatal gavage tube. These extrusions are made from soft, well-tolerated, silicone rubber and incorporate manometric lumina that are smaller in diameter and greater in number than similarly sized polyvinyl assemblies. Previous validation studies have shown that silicone rubber extrusions have acceptable resistance to perfusion and an increase rate of > 100 m m Hg at perfusion rates of 0 . 0 2 5 - 0 . 0 3 mL/min. 14 Polyvinyl assemblies also do not incorporate microsleeve sensors. Unlike side holes, sleeve sensors enable accurate monitoring of LES pressure without the risk of displacement of the sensor from the sphincter with swallowing) la2'18 Previous studies have made estimates of LES pressure using rapid sphincter pull through. These measurements are limited in value because they only give a sample of LES pressure that is unreliable because of the well-documented instability of LES tone over time. 6'19'2° The technical advances described here have enabled the first prolonged study of premature neonates with a perfused sleeve-side hole manometric assembly. In the current study, most esophageal pressure waves were nonperistaltic. Contrary to our expectations, increasing PCA had no effect on the number or percentage

1762

OMARI ET AL.

GASTROENTEROLOGY Vol. 109, No. 6

A

B

pharynxmmHg3°o]~ esophagus I mmHg60o~ ~

C

D

_

~

2 mmHg600 "11

esophagus

stomach

]

mmHg60

0d

.,.-.-.==---.

! 10seconds! Figure 4, Esophageal motility associated with normal swallowing and Santmyer swallowing. In this example, the pharyngeal signal was viable. (A) A normal spontaneous swallow with propagated peristalsis, (B) a Santmyer swallow with propagated peristalsis, (C) a Santmyer swallow that failed to be propagated (peristaltic failure), and (D) failure to stimulate a Santmyer swallow (complete failure). Arrows indicate the point at which the puff of air was administered.

of nonperistaltic contractions apart from retrograde waves. The number of retrograde pressure waves substantially decreased in older infants. There have been conflicting reports on esophageal body motor activity in premature infants. Gryboski= reported that premature infants showed poor propagative peristalsis with a significant proportion (20%) of synchronous pressure waves. A more recent study ~3 suggested esophageal peristalsis was coordinated in infants between 30 and 35 weeks PCA. Nonperistaltic motor patterns are likely to be important in the pathophysiology of reflux, because reflux linked to poor esophageal clearance may lead to pro-

80

0

70

O 0

0

60

0

o

o

50 % SUCCESS

g

0 0

40

0

0

Table 2. LES Pressure

0

30 20

000 oO

10 9

33

0

OOo

O0

o

0

longed esophageal acid exposure and esophageal retention of refluxate, thereby increasing the risk of aspiration and perhaps triggering reflex changes in airway and cardiovascular function.2-4 In adults, peristaltic dysfunction results in impaired esophageal volume clearance] In 1983, Hillemeier et al. 22 described high proportions of nonperistaltic esophageal body contractions in older infants with GER or GER-related morbidity, such as pulmonary disease or failure to thrive. The criteria used by Hillemeier et al. for definition of nonperistaltic motility were given in insufficient detail to allow us to make a competent comparison of our data with theirs. It seems probable that nonperistaltic motility patterns of prematurity are a manifestation of immaturity of the central integration of vagal efferent signals and/or intramural peristaltic control mechanisms. Retrograde pres-

,

34

0

o

q~

3'6

' >37

POST-CONCEPTiONAL AGE

(WEEKS)

Figure 5. The individual data points for relative success of elicitJng Santmyer swallowing in healthy preterm infants 3 3 - 3 7 weeks PCA. Meaned data are shown as bars,

LES pressure (mm Hg) Age group

Preprandial

Postprandial

3 3 - 3 4 weeks (n = 13) 3 5 - 3 6 weeks (n = 12) 3 7 - 3 8 weeks (n = 7)

16.6 _+ 1.6 22.5 + 2.6 25.7 + 5.1

12.9 + 2.3 14.9 + 2.0 a 13.5 + 2.1 a

NOTE. The preprandial and postprandial LES pressures determined for different age groups of healthy premature infants. Data given as mean _+ SEM of n studies, aPostprandial LES pressure significantly less than preprandial LES pressure (P < 0.05),

December 1995

sure waves in particular may reflect a response to lower esophageal distention triggered by reflux or swallowrelated LES relaxation. Relative immaturity of one component of the system may be a factor. Incomplete esophageal pressure waves may result from deficiencies in the cholinergic pathway, which largely determines the amplitude of contractions, while synchronous pressure waves may result from deficiencies in the nonadrenergic, noncholinergic inhibitory pathway that regulates propagation velocity and timing of contractions. Nitric oxide may be the primary mediator responsible for intrinsic descending inhibition of the esophagus during peristalsis. Simultaneous esophageal body contraction has been reported in both the opossum 23 and human 24 esophagus after administration of antagonists of intrinsic NO such as Nm-nitro-L-arginine methyl ester and hemoglobin. Immunohistochemistry of the human esophagus has shown that 36% of myenteric neurons express NO, which is abundant in the circular smooth muscle. 25 The Santmyer reflex was tested because it has been shown to be a reproducible way to elicit swallowing in premature infants. This reflex was described by Orenstein et al. ~5 and is thought to be mediated via sensory somatic fibers of the trigeminal nerve and autonomic efferents of the vagus and glossopharyngeal nerves. ~6 In the current study, swallowing stimulated by the Santmyer reflex was more common in infants whose PCA was > 3 5 weeks compared with younger infants. The importance of increasing PCA on the reflex is, however, unclear in view of the heterogeneity of the response in all age groups. Although it is possible that Santmyer reflex-induced swallowing may to some degree be learned postnatally, other factors such as feed type, body weight, and arousal state did not seem to influence its expression. All infants showed a high-pressure zone at the LES that relaxed with swallowing. One study 9 that used a perfused side hole pull-through technique to record LES pressure found that LES pressure was low in premature infants and in the region traditionally associated with sphincter incompetence ( 0 - 6 m m Hg). Recently, Newell et al. *° found that LES pressure assessed by side hole pull-through increased from 3.8 mm Hg at 29 weeks gestation to 18.1 mm Hg at term. The performance of pull-through may cause physiological disturbance from manometric assembly movement and cannot monitor reflex events such as swallow-induced relaxation. Other studies 8'26 have used an inaccurate nonperfused manometric recording system; these results must be discounted. The present study is the first performed in premature infants that incorporates continuous monitoring of LES pressure with a sleeve sensor. In the current study, LES pressure was higher before

ESOPHAGEAL MOTILITY IN PRETERM NEONATES 1763

feeding and decreased by 30% after feed infusion. A similar decrease in LES pressure has been observed in ambulant subjects after a meal. i9 This may be associated with the change from a fasting to a fed gastrointestinal motor pattern and is possibly mediated by the postprandial increase in plasma cholecystokinin levels. 2v The maturational increase in preprandial basal LES pressure may reflect higher basal plasma cholecystokinin levels in younger infants because of more frequent feeding and/or the maturation of the interdigestive cycle that appears late in gestation, developing fully at 37 weeks PCA. 28 Newell et al. l° similarly reported that the development of LES pressure (recorded immediately before feeding) correlated with PCA and proposed that this indicated maturation of antireflux mechanisms. This seems unlikely, however, given that basal LES pressure is not a major determinant of GER, 6'19'29 and most term infants with or without pathological GER have a competent lEES. 22

Our experience with swallow monitoring using nonperfused, water-filled pharyngeal side holes was disappointing and placed unexpected limitations on our ability to distinguish between primary and secondary esophageal pressure waves. Although the nonperfused side hole was clearly effective in recognizing pharyngeal contraction, maintaining a recognizable swallow signal for the duration of the study proved difficult. Recognition of pharyngeal contraction could have been made more effective by water perfusion. However, we considered it unacceptable to perfuse the pharyngeal side hole because of the possibility of aspiration, especially because newborn and preterm infants have poorly synchronized respiratory inhibition with swallowing. 3°-32 We are currently evaluating alternative approaches to monitoring of swallowing. Despite the problems experienced with swallow recognition in the present study, instances of LES relaxation in response to swallowing were recorded. These indicated that the timing of swallow-induced LES relaxation was similar to that found in older children and adults. A number of LES relaxations independent of swallowing were also observed. Although these seem similar to the transient LES relaxations found in GER in older children, 29 the lack of a pH sensor in the current study limits their interpretation. Relaxation of the LES has been previously recorded in premature infants, 8'1°'26 but the point sensor technique used in these studies does not allow accurate recognition of the timing or completeness of relaxation.~2 We used a novel manometric technique to examine spontaneous and stimulated LES and esophageal body motor function in healthy premature infants. Our data indicate that, although these babies have a functional

1764

OMARI ET AL.

LES, esophageal body motility is largely nonperistaltic when compared with older children and adults. The reduction in esophageal body peristalsis could potentially lead to a failure to clear refluxed material from the esophagus, increasing the risk of subsequent complications. Future studies using the sleeve sensor will be required to define more precisely the mechanisms responsible for GER in these babies.

References 1. Orenstein SR. Gastroesophageal reflux. Curr Probl Pediatr 1991;21:193-241. 2. Herbst J, Minton S, Book L. Gastroesophageal reflux causing respiratory distress and apnoea in newborn infants. J Pediatr 1979;95:763-768. 3. Shaker R, Dodds W, Ren J, Hogan W, Arndorfer R. Esophageal closure reflex: a mechanism of airway protection. Gastroenterology 1992; 102:857-861. 4. Orenstein S, Orenstein D. Gastroesophagel reflux and respiratory disease in children. J Pediatr 1988;112:847-858. 5. Herbst J, Book L, Bray P. Gastroesophageal reflux in the near miss sudden infant death syndrome. J Pediatr 1978; 92:73-75. 6. Dent J, Dodds W, Friedman R, Sekikuchi T, Hogan WJ, Amdorfer RC, Petrie DJ. Mechanism of gastroesophageal reflux in recumbent asymptomatic human subjects. J Clin Invest 1980; 65:256267. 7. Kahrilas P, Dodds W, Hogan W. Effect of peristaltic dysfunction on esophageal volume clearance. Gastroenterology 1988;94: 73-80. 8. Gryboski J. Suck and swallow in the premature infant. Pediatrics 1969;43:96-102. 9. Boix-Ochoa J, Canals J. Maturation of the lower oesophagus. J Pediatr Surg 1976; 11:749-756. 10. Newell S, Sarkar P, Durbin G, Booth I, McNeish A. Maturation of the lower oesophageal sphincter in the preterm baby. Gut 1988; 29:167-172. 11. Dodds W, Stewart E, Hodges D, Zboralske F. Movement of the feline esophagus associated with respiration and peristalsis. J Clin Invest 1973;52:1-13. 12. Dodds W, Stewart E, Hogan W, Stef S, Amdorfer R. Effect of esophageal movement on intraluminal esophageal pressure recording. Gastroenterology 1974; 67:592-600. 13. Kiernan S, Maher K, Benjamin S, DiPalma J. Esophageal motility (EM) in preterm infants (PI) (abstr). Pediatr Res 1992;31:109A. 14. Omari T, Dent J, Bakeweil M, Fraser R, Davidson G. Performance of micro-manometric silicone rubber extrusions at low perfusion rates (abstr). Gastroenterology 1994; 107:1247A. 15. Orenstein S, Giarrusso V, Proujansky R, Kocoshis S. The Santmyer swallow: a new and useful infant reflex. Lancet 1988; 1:345-346. 16. Orenstein S, Bergman I, Proujansky R, Kocoshis S, Giarrusso V.

GASTROENTEROLOGY Vol. 109, No. 6

17.

18. 19.

20. 21. 22.

23. 24.

25.

26. 27.

28.

29. 30.

31.

32.

Novel primitive swallowing reflex: facial receptor distribution and stimulation characteristics. Dysphagia 1992;7:150-154. Hittner H, Hirsch N, Rudolph A. Assessment of gestational age by examination of the anterior vascular capsule of the lens. J Pediatr 1977; 91:455-458. Dent J. A new technique for continuous sphincter pressure measurement. Gastroenterology 1976; 71:263-267. Schoeman MN, Tippitt MD, Akkermans LMA, Dent J, Holloway RH. Mechanisms of gastroesophageal reflux in amulant healthy human subjects. Gastroenterology 1995; 108:83-91. Werlin S, Dodds W, Hogan W, Arndorfer R. Mechanisms ofgastroesophageal reflux in children. J Pediatr 1980;97:244-249. Gryboski J. The swallowing mechanism of the neonate. I. Esophageal and gastric motility. Pediatrics 1965; 36:445-451. Hillemeier AC, Grill BB, McCallum R, Gryboski J. Esophageal and gastric motor abnormalities in gastroesophageal reflux during infancy. Gastroenterology 1983;84:741-746. Anand N, Paterson W. Role of nitric oxide in oesophageal peristalsis. Am J Physiol 1994;266:G123-G131. Murray JA, Conklin JL, Launspach J, Evans D, Ledlow A, Loveday M, Gerber M. Hemoglobin alters esophageal function in humans (abstr). Gastroenterology 1994;107:1228A. Singaram C, Sengupta A, Sweet M, Sugerbaker D, Goyal R. Nitrinergic and peptidergic innervation of the human oesophagus. Gut 1994; 35:1690-1696. Gryboski J, Thayer W, Spiro H. Esophageal motility in infants and children. Pediatrics 1963;31:382-395. Ciaccia D, Brazer S, Lawson D, Mantyh C, Liddle R, Pappas T. Cholecystokinin is a physiological mediator of fat-induced lower esophageal sphincter relaxation (abstr). Gastroenterology 1995; 108:A584. Bisset W, Watt J, Rivers R, Milla P. Ontogeny of fasting small intestinal motor activity in the human infant. Gut 1988; 29:483488, Davidson G, Dent J, Willing J. Monitoring of upper oesophageal sphincter pressure in children. Gut 1991;32:607-611. Bamford O, Taciak V, Gewolb IH. The relationship between rhythmic swallowing and breathing during suckle feeding in term neonates. Pediatr Res 1992;31:619-624. Koenig JS, Davies AM, Thach BT. Coordination of breathing, sucking, and swallowing during bottle feedings in human infants. J Appl Physiol 1990;69:1623-1629. Mathew OP, Bhatia J, Sucking and breathing patterns during breast- and bottle-feeding in term neonates. AJDC 1989;143: 588-592.

Received May 30, 1995. Accepted August 16, 1995. Address requests for reprints to: Taher I. Omari, Ph.D., Department of Gastroenterology, Women's and Children's Hospital, North Adelaide, Australia 5006. Fax: (61) 8-204-6088. Supported by the National Health and Medical Research Council of Australia and by the Adelaide Women's and Children's Hospital Foundation.