Browse by Author
Browse by Movies
How to separate the spitters from the serious refluxers in infants?
S.G. Jolley (Las Vegas)
The task of separating the normal "spitting" infant from the infant with serious disease from emesis is not as straightforward as one might presume. Regurgitation of feedings may be effortless or with effort. The material regurgitated may have bile or be nonbilious. Several causes for "excessive" emesis can be found in infants, but the main goal of diagnosis is to exclude obstruction in the alimentary tract as the cause for emesis. Forceful emesis and emesis with bile are more likely to be indicative of alimentary tract obstruction (e.g., pyloric stenosis, intestinal malrotation, intussusception) than is effortless emesis without bile. Since the diagnostic features of alimentary tract obstruction in infants are well described in pediatric surgical texts, this article will focus primarily on excessive emesis in infants which is nonbilious and effortless.
Historically, the barium meal has been used as an initial test for the evaluation of infants with nonbilious, effortless emesis [1,2]. The study is most suited to exclude the presence of anatomical abnormalities which may produce obstruction in the foregut, or may be associated with emesis (e.g., hiatal hernia). Since reflux of barium from the stomach into the esophagus occurs in normal infants [3] and may be absent in infants with serious disease from emesis [3,4], the presence of reflux during the barium meal is not a reliable method for separating infants with disease from the normal "spitty" infant. A radionuclide esophagram [5] is easier to quantitate, but no more accurate than the barium meal for identifying infants with disease from emesis.
Manometric measurements of the esophagus and stomach also have been employed to identify infants with disease resulting from emesis as a consequence of a defective gastroesophageal junction (gastroesophageal reflux disease (GERD)) [6]. Intuition would suggest that infants with a defective gastroesophageal junction causing disease should have a decreased lower esophageal sphincter pressure (LESP). In fact, it is not that simple. Infants with GERD can have a decreased, normal or increased LESP [6]. The dynamic function of the LES is more important than static measurements and shows inappropriate relaxations of the LES or a persistent decreased basal LESP as associated with a defective gastroesophageal junction in infants with GERD [7]. Therefore, a single static measurement of LESP cannot reliably identify the presence or absence of a defective gastroesophageal junction in infants. Prolonged manometric studies in infants are just evolving and will require a critical analysis of data from normal infants to establish a standard reference.
Endoscopy has been used to identify esophageal mucosal abnormalities as an indicator of disease from excessive emesis in infants. The presence of mucosal ulceration, stricture, extreme friability or Barrett's epithelium are clearly evidence of mucosal injury, probably from GERD. However, most infants with GERD do not have these signs of advanced mucosal injury. Furthermore, less obvious findings of mucosal injury on microscopic examination of the esophageal biopsy in infants
Figure 1. .A comparison of normal infants and children (control, n = 24) and symptomatic infants and children (symptomatic, n = 12) with well documented gastroesophageal reflux disease having 24-h esophageal pH monitoring perfomed and analyzed by a) the Johnson-DeMeester score; b) the percentage of total time that the esophageal pH < 4; and c) the author's method using an esophageal pH score calculated from recording intervals after the first 2 h postcibal. The dotted lines represent the upper limit of normal for the children studied [13].
[8-10] may result from any mechanism (including alimentary tract obstruction) whereby excessive gastric contents are regurgitated into the esophagus and GERD is not present.
Due to the dissatisfaction with the barium meal, manometric methods and endoscopy (in other than severe forms of GERD) to identify disease from excessive emesis, extended esophageal pH monitoring became popular in the 1970s. The methodology utilized by Johnson and DeMeester in adults [11] has been tried in infants and children [12] but has been less successful in identifying infants with GERD than the success reported for adult studies [13]. In 1977, the author attempted to apply the method of Johnson and DeMeester to a group of normal infants and children [13,14]. Unfortunately, the method of Johnson and DeMeester produced a 50% false-negative rate in infants with documented GERD (Fig. la). When the percentage of total time esophageal pH < 4, the most commonly reported parameter to indicate GERD, was used, the false-negative rate was 50% and the false-positive rate was 33% (Fig. 1b).
Figure 2. .An algorithm used by the author to separate infant spitters from infants with disease caused by excessive emesis.
The reasons behind the low accuracy with the Johnson-DeMeester scoring method, or the use of the percentage of total time with esophageal pH < 4 in infants, were due to these methods of analysis ignoring the basic stimuli for acid reflux episodes in normal children, such as feedings and the state of wakefulness [14,15]. Another method for esophageal pH record analysis devised by the author accounted for both feeding and wakefulness periods separately, thereby resulting in an esophageal pH score which more reliably identifies infants and children with GERD (Fig. 1c). The false-negative rate with the author's method is less than 5%. Although GERD in infants occurs in various patterns (type I, II and III) [16], the method for deriving the pattern type cannot replace the esophageal pH score for identifying GERD in infants. The reflux patterns in infants serve to characterize the GERD rather than to confirm its existence.
The infants with disease resulting from excessive emesis may have gastric emptying abnormalities in addition to GERD, or as the sole cause for the emesis. A radionuclide gastric emptying study with a clear liquid, and mathematically corrected for delay caused by postcibal gastroesophageal reflux episodes, can identify infants with abnormal effective gastric emptying [17]. The corrected gastric emptying may be slow, normal or rapid. A slow corrected gastric emptying is seen in 25% of symptomatic infants with GERD and in 29% without GERD. The slow corrected gastric emptying is a separate cause of disease from excessive emesis in infants. Rapid corrected gastric emptying usually causes dumping symptoms which occasionally are associated with excessive emesis and no GERD in infants [17].
The author's preferred method for identifying infants with disease from excessive emesis is shown in Fig. 2. As mentioned previously, the main goal is to exclude anatomical obstruction. Once that has been accomplished, 18-24-h esophageal pH monitoring and radionuclide gastric emptying studies are important in the symptomat-
Table 1. .The combination of findings from esophageal pH monitoring and gastric emptying studies in infants with excessive emesis by history and no indication of alimentary tract obstruction by physical examination and radiographic studies
|
Esophageal pH score |
Pattern type |
Corrected gastric emptying |
Incidence (%) |
Interpretation |
||
|
Abnormal |
I |
Slow |
14 |
GERD unlikely to resolve & delayed gastric emptying |
||
|
|
I |
Control |
17 |
GERD unlikely to resolve |
||
|
|
I |
Fast |
24 |
GERD unlikely to resolve and rapid gastric emptying |
||
|
|
II |
Slow |
7 |
GERD likely to resolve & delayed gastric emptying |
||
|
|
II |
Control |
5 |
GERD likely to resolve |
||
|
|
II |
Fast |
11 |
GERD likely to resolve and rapid gastric emptying |
||
|
|
III |
Slow |
1 |
GERD unlikely to resolve & delayed gastric emptying |
||
|
|
III |
Control |
3 |
GERD unlikely to resolve |
||
|
|
III |
Fast |
5 |
GERD unlikely to resolve and rapid gastric emptying |
||
|
Normal |
Normal |
Slow |
4 |
Delayed gastric emptying |
||
|
|
Normal |
Control |
8 |
Normal |
||
|
|
Normal |
Fast |
1 |
Rapid gastric emptying |
||
ic infant to assess for GERD and abnormalities of effective gastric emptying. Virtually any combination of findings is possible (Table 1), but the abnormalities must be treated separately. Regurgitation of feedings due to poor esophageal emptying has been difficult to quantitate in infants. A delay in esophageal emptying may be suspected by either grossly delayed emptying of barium from the esophagus during a barium meal [18], or the prolonged clearance of apple juice feedings from the esophagus during extended esophageal pH monitoring [19].
In summary, the author's method for performing extended esophageal pH monitoring is the most accurate method currently available to identify infants who have GERD as a cause for excessive emesis. A corrected gastric emptying value from liquid radionuclide gastric emptying studies also allows the clinician to identify infants with excessive emesis secondary to gastric emptying abnormalities, regardless of whether or not GERD is also present.
References