Introduction

Gastroesophageal reflux (GER), the involuntary passage of gastric contents into the esophagus, occurs several times per day in every human, particularly after meals, and is a completely normal physiologic process. Most reflux episodes are asymptomatic, of short duration, and limited to the distal esophagus. Regurgitation—also called spitting-up, posseting, or spilling—is the passage of GER into the pharynx, mouth, or the perioral external area. Regurgitation is frequent in healthy infants, especially in the first months of life, with a peak incidence around 3 to 4 months, and after intake of large volumes as happens in young infants. Vomiting is a forceful expulsion of gastric contents from the mouth. Vomiting is a more complex phenomenon caused by a coordinated autonomic and voluntary motor response as a consequence of the activation of receptors, which are often located external to the gastrointestinal (GI) tract. GER disease (GERD) occurs when GER causes troublesome symptoms and/or complications. The spectrum of GER symptoms is atypical and nonspecific (such as behavioral, respiratory, feeding, or sleeping problems with or without esophageal signs and symptoms such as regurgitation or vomiting). GERD frequently causes an impaired quality of life, which is determined mainly by parental perception and coping during infancy and early childhood.

Mucosal complications of GERD such as Barrett’s esophagus are probably less frequent in children than in adults. The natural history, evolution, and progression of GERD are unclear for an individual patient and may depend on genetic, environmental, and mucosal factors.

In 2009, combined European and North American guidelines on the diagnosis and management of GER in children were published. In 2013 the American Academy of Pediatrics (AAP) confirmed and approved this document. However, it is difficult to inform general or family pediatricians about these guidelines. A consequence of the latter is that overdiagnosis of GERD and thus overtreatment with medication (mainly acid inhibitors) is common. Children are frequently prescribed proton pump inhibitors (PPIs) based on a previous (mis)diagnosis of GERD.

Prevalence

Determination of the exact prevalence of GER and GERD at any age is nearly impossible, mainly because symptoms are not specific, not all patients seek medical help, and many patients are not (fully) investigated. Many epidemiologic studies have evaluated the frequency of regurgitation. It is often postulated erroneously that GER and GERD are more frequent in infants than in children and adults; these data are not available. Regurgitation is more frequent in infants between 1 and 6 months of age than in any other age group. Symptoms of acid regurgitation, heartburn, or both, at least once a week occur in 10% to 20% of adults in the Western world, whereas in Asia the prevalence is roughly less than 5%.

In normal 3- to 4-month old infants, three to four episodes of GER are detectable during 5 minutes of intermittent fluoroscopic evaluation. According to esophageal pH monitoring, up to 31 ± 21 acid reflux episodes are recorded within a 24-hour period in infants. More recently, studies using esophageal impedance have reported up to 100 episodes and 70 episodes of reflux in 24 hours in infants and children, respectively. How­ever, for ethical reasons, these investigations were performed in symptomatic children. Less than 10% of infants and children have (acid and troublesome) GERD.

The rapidly increasing prevalence of obesity is causing a rising prevalence of GERD. The risk of GERD symptoms is associated with the increase in body mass index and waist circumference, even in normal-weight children. Total and abdominal obesity are risk factors for GERD symptoms in children.

Pathophysiology

Inappropriate transient lower esophageal sphincter relaxation (TLESR) is the most important pathophysiologic mechanism causing GER at any age. A TLESR is normal following a swallow; it is a neural reflex, triggered mainly by the distension of the proximal stomach and organized in the brainstem, with efferent and afferent pathways traveling in the vagus nerve, activating an intramural inhibitory neuron that releases nitric oxide to relax the lower esophageal sphincter (LES). TLESR is the main mechanism underlying GER events in patients with esophageal atresia (EA). Most infants and adults have impaired motility, delayed bolus clearance, and delayed gastric emptying. Increased abdominal or decreased LES basal pressure is less commonly involved.

GER is influenced by genetic, environmental (e.g., diet and smoking), anatomic, hormonal, and neurogenic factors ( Figure 21-1 ). Three major lines of defense limit the degree of GER and GERD: the anatomic “antireflux barrier,” consisting of the LES and the diaphragmatic pinchcock and angle of His, the esophageal peristalsis and clearance, and the esophageal mucosal resistance. Esophageal mucosa defense can be divided in preepithelial (protective factors in saliva and esophageal secretions containing bicarbonate, mucin, prostaglandin E2, epidermal growth factor, and transforming growth factor), epithelial (tight junctions and intercellular glycoprotein material), and postepithelial factors. Interindividual variation of reflux perception suggests different esophageal sensitive thresholds, which is in part determined by capsaicin levels and vanilloid receptor-1 activity There are acid-, temperature-, and volume-sensitive receptors in the esophageal mucosa. Widened intercellular spaces are found in patients with esophagitis and endoscopy-negative disease. Esophageal sensitivity to acid decreases when the esophagitis has healed. Duodenal fat increases the sensitivity to reflux.

Pathophysiologic mechanisms for GER.

Gastric distension and impaired accommodation of the gastric fundus may exacerbate TLESRs. Delayed gastric emptying may increase postprandial reflux, possibly by increasing the rate of TLESRs. Delayed gastric emptying has been documented in (a proportion of) infants and children with symptomatic GER, in particular in those with neurologic disorders. Esophageal acid exposure in patients with GERD is directly correlated with the emptying time of the proximal stomach. Acidity of the refluxate may also relate to a localized proximal gastric area called “the gastric acid pocket,” which may persist even in the postprandial period when (the rest of the) stomach content is neutralized by the meal. Position and sleeping influence GER and gastric emptying. In the recumbent position, noxious gastric materials, rather than air, are positioned at the cardia and may move more easily into the esophagus, especially when the LES tone is decreased during sleep. Both salivation and swallowing are markedly reduced during sleep, further impairing clearance.

Hiatal hernia increases the number of reflux episodes, and delays esophageal clearance by promoting retrograde flow across the esophagogastric junction when the LES relaxes after a swallow. This mechanism underlies the so-called re-reflux phenomenon (acid reflux when the pH is still below 4).

Carre described autosomal dominant inheritance of hiatal hernia by discovering familial hiatal hernia in five generations of a large family, but without demonstrating the link to GERD. Moreover, the concordance for GER is higher in monozygotic than dizygotic twins. Genes in question have been localized to chromosomes 9 and 13. A locus on chromosome 13q, between microsatellite D13S171 and D13S263, has been linked with severe GERD in five multiply affected families. This could not be confirmed in another five families, probably because of the genetic heterogeneity of GERD and different clinical presentations of patients. The relevance of these findings for the general population remains unclear.

Most pathophysiologic mechanisms of GERD have been studied in adults, and acid reflux has long been considered as the critical factor. However, in the last decade, an increasing body of evidence (based on esophageal impedance combined with pH monitoring) has shown that weakly acid reflux (that is GER with a pH between 4 and 7) is equally associated with symptoms, particularly in infants, during the postprandial period and in patients who are resistant to acid inhibitors. Components con­tributing to the noxiousness of refluxed material are pepsin, bile acids and salts, and trypsin. The latter two depend on duodenogastric reflux preceding GER and are implicated in the genesis of strictures and Barrett’s esophagus.

In respiratory symptoms related to GER, both (micro)aspiration and neurally (vagally) mediated mechanisms have been considered.

The role of Helicobacter pylori infection and the benefit of eradication of H. pylori in patients with GERD have long been debated without a clear conclusion.

Symptoms and Signs

Although reflux does occur physiologically at all ages, there is at all ages also a continuum between physiologic GER and GERD leading to significant symptoms, signs, and complications ( Boxes 21-1 and 21-2 and Table 21-1 ). GER symptoms are more frequent in children with functional constipation. The presenting symptoms of GERD differ according to age.

Belching or eructation occurs during TLESR, and is an important method of venting air from the stomach. Hiccups are involuntary reflex contractions of the diaphragm followed by laryngeal closure. In some cases, hiccups cause GER.

Atypical symptoms such as epigastric pain, nausea, flatulence, hiccups, chronic cough, asthma, chest pain, and hoarseness account for 30% to 60% of presentations of GERD. Possible associations exist between GERD and asthma, pneumonia, bronchiectasis, ALTE (acute life-threatening event), laryngotracheitis, sinusitis, and dental erosion, but causality or temporal association have not been established. The paucity of studies, small sample sizes, and varying disease definitions do not allow for firm conclusions.

Clinicians need to be aware that not all regurgitation and vomiting in infants and young children is related to GER/GERD. Bilious vomiting, gastrointestinal bleeding, consistently forceful vomiting, weight loss or failure to thrive, diarrhea, constipation, fever, lethargy, hepatosplenomegaly, abdominal tenderness, and/or distension should raise the possibility of an alternate diagnosis.

Diagnosis

Detailed information regarding indications and pitfalls of radiologic contrast studies, nuclear reflux scintig­raphy, ultrasound, pH-metry and intraluminal impedance, endoscopy, manometry, gastric emptying tests, and electrogastrography can be found in review papers and guidelines.

In adults, diagnosis of GERD is based primarily on clinical history. However, even in adults, the sensitivity and specificity of history is questioned. History is useful for predicting acid GERD, but not for diagnosing nonacid GERD, and is inferior to MII-pH. However, in children younger than 8 years of age, or even younger than 12 years of age, history is considered poorly reliable. Questionnaires were developed to improve the reliability of history. Orenstein developed the Infant Gastroesophageal Reflux Questionnaire (I-GERQ), which results in an objective, validated, and repeatable quantification of symptoms suggestive for GERD. The I-GERQ was revised (the I-GERQ-R) in 185 patients and 93 controls, resulting in an internal consistency and test–retest reliability of greater than 0.85. However, Aggarwal and co-workers obtained with the same I-GERQ a sensitivity of only 43% and a specificity of 79%. Moreover, pH-metry results were not different according to a “positive” or “negative” score of the I-GERQ. Vandenplas and co-workers showed that not one question was found to be significantly predictive of the presence of esophagitis; I-GERQ cut-off score failed to identify 26% of infants with GERD (according to pH-metry results or presence of esophagitis) and was positive in 81% of infants with a normal esophageal histology and normal pH-metry results. Deal et al. developed two different questionnaires, one for infants and one for older children, and showed that the score was higher in symptomatic than in asymptomatic children. In other words: the correlation between the results of history obtained with the help of questionnaires on the one side and of reflux investigations on the other side, is poor. Because the existing questionnaires seem to be of limited value, more research is needed on this topic. In many countries, history is the only diagnostic tool available, and at any rate one should have a reliable history to select the appropriate children to investigate.

Barium contrast radiography, nuclear scintiscanning, and ultrasound are techniques that evaluate postprandial reflux and provide also some information on gastric emptying. Normal ranges are not established for any of these procedures. There is broad consensus that barium studies are not recommended as first-line investigation for diagnosis of GERD, but radiologic examination is of importance for diagnosis of anatomic abnormalities such as malrotation, duodenal web, and stenosis, and may suggest functional abnormalities such as achalasia. In addition, in many countries, barium contrast is still the only investigative technique available. Nuclear scintigraphy may show pulmonary aspiration. However, these data still need confirmation, since experience from many centers indicates that aspiration occurs only seldom. Aspiration of saliva and gastric contents occurs physiologically during sleep in healthy adults. Scintigraphy also evaluates gastric emptying. But, the 13C-octanoic acid (for solids) and 13C-acetate (for liquids) breath tests are more appropriate for measuring gastric emptying. The role of delayed gastric emptying in GER(D) remains controversial. The results of ultrasound are investigator dependent, and a relation between reflux seen on ultrasound and symptoms has not been established. There is no indication for electrogastrography in the diagnostic workup of a patient suspected of GERD.

Modern endoscopes are so miniaturized that scoping of preterm infants weighing less than 1,000 g has become technically easy. Operator experience is an important component of interobserver reliability. Endoscopy allows direct visual examination of the esophageal mucosa. Macroscopic lesions associated with GERD include esophagitis, erosions, exudate, ulcers, structures, and hiatal hernia. Redness of the distal esophagus in young infants is a normal observation because of the increased number of small blood vessels in the cardiac region. Endoscopy may also show a “sliding hernia,” the stomach that is protruding into the esophagus during burping. Recent consensus guidelines define reflux esophagitis as the presence of endoscopically visible breaks in the esophageal mucosa at or immediately above the gastroesophageal junction. Endoscopy-negative reflux disease is common. There is a poor correlation between the severity of symptoms and presence and absence of esophagitis. There is insufficient evidence to support histology to diagnose or exclude GERD. Dilated intercellular space diameter seems to be a useful and objective marker for esophageal damage in pediatric GERD, regardless of acid exposure. In childhood, different GERD phenotypes cannot be discriminated on the basis of reflux pattern. In children with reflux-related cough, dilated intercellular space diameter appears to be an objective and useful marker for esophageal mucosal injury regardless of acid exposure, and its evaluation should be considered for those patients in whom the diagnosis is uncertain. Biopsies of duodenal, gastric, and esophageal mucosa are mandatory to exclude other diseases. More detailed information on pros and cons of histology can be found in consensus papers.

Manometry does not demonstrate reflux, but it is of interest to analyze pathophysiologic mechanisms causing the reflux, mainly by visualizing and measuring TLESRs, and is indicated in the diagnosis of specific conditions such as achalasia. Ambulatory 24-hour esophageal manometry, in combination with pH-metry and/or impedance recording, is nowadays technically feasible. This technique is used mainly in (clinical) research, and allows the objective demonstration of reflux-symptom association (e.g., in patients presenting with chronic cough).

Experience in children with spectrophotometric esophageal probes to detect bilirubin remained limited. Using these probes, Orel and co-workers showed that some children with esophagitis have bile reflux. However, because of technical limitations, the technique is not applied routinely.

Esophageal pH-metry is the best method for measuring acid in the esophagus, but not all reflux that causes symptoms is acidic and not all acid reflux causes symptoms. Esophageal pH-metry is useful for evaluating the effect of a therapeutic intervention on reducing esophageal acid exposure. Normal ranges have been established for pH-metry. However, results depend on the hardware and software used. Normal ranges have been established using glass electrodes, which have been shown (in several studies in children and in adults) to pick up more acid reflux episodes than antimony electrodes, which are nowadays the most popular. Esophageal pH monitoring is likely to be of limited value for reflux causing extraesophageal manifestations. The demonstration of a time association between GER episodes and symptoms is one of the major indications for this technique. The most common MII-pH abnormality was an abnormal symptom association between cough and reflux and the most common endoscopic abnormality was reflux esophagitis. However, the optimal time-frame to be considered as “time-association” and the optimal parameter to calculate significant association are still debated. Oropharyngeal pH-metry detected fewer reflux episodes than MII-pH; 35% of the oropharyngeal events were swallows according to MII-pH. Time correlation between cough and reflux could not be demonstrated with oropharyngeal pH-metry.

MII measures electrical potential differences. As a consequence, the detection of reflux with MII is not pH dependent, but in combination with pH-metry MII allows detection of acid (pH <4.0), nonacid, or weakly acidic (pH 4.0-7.0) and alkaline reflux (pH > 7.0). MII in combination with pH-metry definitively measures more reflux episodes than pH-metry alone. MII-pH in newborns has prognostic value regarding the duration of GERD symptoms and provides useful information that clinicians may share with parents about the prognosis of symptomatic infants. Impedance bolus exposure index and proximal reflux frequency seem to be the variables with the highest predictive value. Weakly acidic reflux events play an important role in determining the duration of GERD symptoms in newborns. MII is more sensitive than pH monitoring for establishing the diagnosis of GER in nonenterally fed critically ill children with mechanical ventilation in the first 48 hours after admission to the pediatric intensive care unit (PICU) because the refluxate is an alkaline or weak acid in the majority of episodes. Patients who received muscle relaxants had a lower frequency of GER. Of interest, pH-only episodes, that is, reflux episodes detected with pH-metry but not with impedance (drop in pH without bolus movement) occur in young children. pH-only events occur mainly during the night and in young infants. Whether this is the “infant presentation” of the “gastric acid pocket,” which was recently described in adults with symptomatic reflux that is resistant to antacid medications is not known. Impedance also measures gas reflux, since liquid causes a drop in impedance and an increase in gas. In adults, baseline impedance was shown to be lower in esophagitis, and treatment of esophagitis with PPIs does increase baseline impedance. The sensitivity and specificity of baseline impedance as a marker for esophageal injury is debated in the literature. In children with reflux-related cough, baseline impedance levels were shown to have no role in identifying reflux-induced esophageal mucosal ultrastructural changes. Pathologic MII-pH results are not predictive for an erosive esophagitis and vice versa. A lower baseline impedance is associated with esophagitis, LA-grade B or higher and may be caused by longer acid exposure, but also by longer bolus exposure, and thus may be another useful parameter in GERD monitoring. These differences in observation regarding the utility of determining the impedance baseline may be related to the baseline impedance, which is age-dependent. The presence of esophagitis may also determine a decreased impedance baseline regardless of the age of the patients. The role of nonacid reflux measured by impedance pH testing in adult patients presenting with symptoms that were refractory to PPI treatment is uncertain at best, and the results from this test alone should not be used to refer patients for surgical fundoplication.

In a substantial subgroup of (adult) patients with heartburn that responded to PPIs, diagnosis of GERD could not be confirmed with MII-pH monitoring. This could mean that PPI response and heartburn are not reliable predictors of GERD diagnosis, or that MII-pH recording is not (always) a reliable diagnostic technique. Acid-related parameters are significantly related to MII baselines. An excellent correlation between manual and automated analyses of MII baselines was found. Distal MII baselines were significantly lower in children with a positive overall MII-pH outcome compared with the proximal esophagus.

Interpretation of an impedance recording is laborious and necessitates experience, since the automatic analysis is not yet standardized and validated in (young) children and infants. One of the major clinical interests of MII seems to be demonstration of symptom association, but normal data and validation of symptom association parameters in children are missing. Recently, attention has focused on baseline impedance, which is age dependent, decreased in esophagitis, and increased with PPI treatment. Infants (especially in the first months of life) and young children present a significantly lower impedance baseline compared to older children, both in the proximal and distal esophagus. Given the high cost of equipment, electrodes, the time needed for analysis and interpretation of impedance, the pros and cons in comparison to pH-metry are still debated. As long as medical therapeutic options are limited primarily to antacid medication, the clinical impact of measuring nonacid reflux is poor. In the meantime, research should focus on determining whether nonacid or weakly acid reflux is causally associated with symptoms.

Indirect techniques have been developed, mainly to diagnose GER(D) in patients with extraesophageal manifestations. Accumulation of evidence regarding the determination of lipid laden macrophages (LLMs) in bronchoalveolar liquid (BAL) resulted in the conclusion that this method lacks sensitivity and specificity. More recent data show the presence of pepsin and bile in BAL and middle ear fluid. There is no prospective, double-blind, placebo-controlled study treating reflux and evaluating the ear, nose, and throat (ENT) outcome as primary endpoint. It has yet to be determined if this association, if any, is “causal” or “consequence.”

An empiric trial of PPIs in adult patients without alarm features can be helpful, but the response to therapy is variable. Empirical treatment in children may be an even bigger challenge, since children are less accurate in reporting symptoms.

In conclusion, all GER-investigation techniques measure different aspects of reflux. Therefore, it is not unexpected that the correlation between the results of the different techniques is poor. There is no “always-best” investigation technique for diagnosing GER(D) because the clinical situation of each individual patients differs. Endoscopy is the only diagnostic tool to identify esophagitis; 24-hour pH-metry measures acid GER, MII detects acidic and weakly acidic GER episodes and reflux-associated symptoms. Barium meal is the best technique to provide information about the anatomy, and is in many countries the only technique available, although it has been shown to be unreliable for diagnosing GERD.